KR20170044660A - Methods of determining tissues and/or cell types giving rise to cell-free dna, and methods of identifying a disease or disorder using same - Google Patents

Abstract

The present disclosure provides a method for determining one or more tissues and / or cell types that contribute to acellular DNA ("cfDNA") in a biological sample of a subject. In some embodiments, the disclosure provides a method of identifying a disease or disorder in a subject as a function of one or more determined tissues and / or cell types that contribute to cfDNA in a biological sample of the subject.

Description

FIELD OF THE INVENTION The present invention relates to a method for determining a tissue and / or cell type that produces a cell-free DNA, and a method for identifying a disease or disorder using the same, and a method for identifying a disease or disorder using the same. DISEASE OR DISORDER USING SAME}

Priority claim

This application claims the benefit of U.S. Provisional Patent Application No. 62 / 029,178, filed July 25, 2014, and U.S. Provisional Patent Application No. 62 / 087,619, filed December 4, 2014, each of which is incorporated herein by reference in its entirety as if fully set forth herein. Based on the following.

Mention of US Government Rights

The present invention was made with government support under grant number 1DP1HG007811 granted by the National Institutes of Health (NIH). The US Government has certain rights to the invention.

Technical field

The present disclosure relates to methods for determining one or more tissues and / or cell types that produce acellular DNA. In some embodiments, the disclosure provides a method of identifying a disease or disorder in a subject as a function of one or more determined tissues and / or cell types associated with acellular DNA in a biological sample of the subject.

Cell-free DNA ("cfDNA") is present in circulating plasma, urine and other human fluids. cfDNA contains a double-stranded DNA fragment that is relatively short (predominantly less than 200 base pairs) and generally low in concentration (e.g., 1-100 ng / mL in plasma). In circulating plasma of healthy individuals, cfDNA is thought to originate primarily from the apoptosis of blood cells (i.e. normal cells of the hematopoietic lineage). However, in certain circumstances, other tissues may contribute substantially to the composition of cfDNA in body fluids such as circulating plasma.

Although cfDNA has been used in certain specialties (e.g., reproductive medicine, cancer diagnosis and transplant medicine), existing tests based on cfDNA have been used in more than one cell population (e. g., maternal to fetal genomes, The genome, the recipient genome to the donor genome, etc.) (e.g., the primary sequence or copy number of a particular sequence). Unfortunately, the overwhelming majority of the cfDNA fragments found in any given biological sample originate from regions of the same genome between the contributing cell populations, so existing cfDNA-based tests are of limited application. In addition, many diseases and disorders involve changes in tissue and / or cell types that produce cfDNA, for example, from inflammatory processes associated with tissue damage or disease or disorder. Existing cfDNA-based diagnostic tests that depend on differences in primary sequence or copy number presentation of a particular sequence between two genomes can not detect this change. Thus, the potential of cfDNA to provide a robust non-biopsy diagnostic method is very large, but there is still a need for a cfDNA-based diagnostic method that can be applied to diagnose a wide variety of diseases and disorders.

The present disclosure provides methods for determining one or more tissues and / or cell types that produce acellular DNA ("cfDNA") in a biological sample of a subject. In some embodiments, the disclosure provides a method of identifying a disease or disorder in a subject as a function of one or more determined tissues and / or cell types associated with cfDNA in a biological sample of the subject.

In some embodiments, the disclosure provides a method of determining the tissue and / or cell type that produces acellular DNA (cfDNA) in a subject, the method comprising isolating cfDNA from a biological sample from the subject, wherein the cfDNA comprises a plurality of cfDNA fragments; Determining a sequence associated with at least a portion of a plurality of cfDNA fragments; determining the genomic location in the reference genome for at least a portion of the cfDNA fragment endpoints of the plurality of cfDNA fragments as a function of the cfDNA fragment sequence; And determining at least a portion of the tissue and / or cell type that produces the cfDNA fragment as a function of the genomic location of at least a portion of the cfDNA fragment endpoint.

In another embodiment, the disclosure provides a method of identifying a disease or disorder in a subject, the method comprising isolating cell-free DNA (cfDNA) from a biological sample from the subject, wherein the isolated cfDNA comprises a plurality of cfDNA fragments Comprising; Determining a sequence associated with at least a portion of a plurality of cfDNA fragments; determining the genomic location in the reference genome for at least a portion of the cfDNA fragment endpoints of the plurality of cfDNA fragments as a function of the cfDNA fragment sequence; determining at least a portion of the tissue and / or cell type that produces cfDNA as a function of genomic location of at least a portion of the cfDNA fragment endpoint; And identifying the disease or disorder as a function of the determined tissue and / or cell type producing the cfDNA.

In another embodiment, the disclosure provides a method of determining the tissue and / or cell type that produces acellular DNA (cfDNA) in a subject, the method comprising: (i) obtaining a biological sample from the subject, isolating cfDNA, constructing a library of cfDNA, and measuring distribution (a), (b) and / or (c) by massively parallel sequencing to generate a nucleosome map; (a), (b) and / or (c) by obtaining a biological sample from a control subject or a subject with a known disease, isolating the cfDNA from the biological sample, and constructing the library of cfDNA and mass- To produce a reference set of nucleosome maps; And (iii) comparing the nucleosome map derived from cfDNA from the biological sample with a reference set of nucleosome maps to determine the tissue and / or cell type that produces cfDNA from the biological sample; (A) the distribution of the likelihood that any particular base pair in the human genome will appear at the end of a cfDNA fragment; (b) the distribution of the likelihood that any pair of base pairs of the human genome will appear as the ends of a pair of cfDNA fragments; And (c) the distribution of the likelihood that any particular base pair in the human genome will appear within the cfDNA fragment as a result of differentially occupied nucleosomes.

In another embodiment, the disclosure provides a method of determining the tissue and / or cell type that produces cfDNA in a subject, the method comprising: (i) obtaining a biological sample from the subject, isolating the cfDNA from the biological sample (a), (b) and / or (c) by constructing a library of cfDNA and massively parallel sequencing to prepare a nucleosome map; (ii) obtaining a biological sample from a control subject or a subject having a known disease, isolating the cfDNA from the biological sample, and isolating the DNA from the fragmentation of the chromatin using an enzyme such as micrococose nuclease, DNase or transporter Measuring distribution (a), (b) and / or (c) by library construction and massively parallel sequencing to create a reference set of nucleosome maps; And (iii) comparing the nucleosome map derived from cfDNA from the biological sample with a reference set of nucleosome maps to determine the tissue and / or cell type that produces cfDNA from the biological sample; Wherein (a), (b) and (c) are as follows: (a) the distribution of the likelihood that any particular base pair in the human genome will appear at the end of the fragment to be sequenced; (b) the distribution of the likelihood that any pair of base pairs of the human genome will appear as the ends of a pair of sequenced fragments; And (c) the distribution of the likelihood that any particular base pair in the human genome will appear in the sequenced fragment as a result of discriminatory nucleosomal occupancy.

In another embodiment, the disclosure provides a method of diagnosing a clinical condition in a subject, the method comprising: (i) obtaining a biological sample from the subject, isolating cfDNA from the biological sample, constructing a library of cfDNA, Measuring a distribution (a), (b) and / or (c) by a crystal to prepare a nucleosome map; (a), (b) and / or (c) by obtaining a biological sample from a control subject or a subject with a known disease, isolating the cfDNA from the biological sample, and constructing the library of cfDNA and mass- To produce a reference set of nucleosome maps; And (iii) comparing the nucleosome map derived from the cfDNA from the biological sample with a reference set of nucleosomal maps to determine the clinical condition; (A) the distribution of the likelihood that any particular base pair in the human genome will appear at the end of a cfDNA fragment; (b) the distribution of the likelihood that any pair of base pairs of the human genome will appear as the ends of a pair of cfDNA fragments; (c) The distribution of the likelihood that any particular base pair in the human genome will appear within the cfDNA fragment as a result of differentially occupied nucleosomes.

In another embodiment, the disclosure provides a method of diagnosing a clinical condition in a subject, the method comprising: (i) obtaining a biological sample from the subject, isolating the cfDNA from the biological sample, and constructing a library of cfDNA and a mass parallel sequence Measuring a distribution (a), (b) and / or (c) by a crystal to prepare a nucleosome map; (ii) obtaining a biological sample from a control subject or a subject having a known disease, isolating the cfDNA from the biological sample, and isolating the chromosome from the chromatin fragmentation using an enzyme such as micrococose nuclease (MNase), DNase or transporter Measuring a distribution (a), (b) and / or (c) by constructing a library of DNA and mass parallel sequencing to generate a reference set of nucleosome maps; And (iii) comparing the nucleosome map derived from cfDNA from the biological sample with a reference set of nucleosome maps to determine the original tissue composition of the cfDNA from the biological sample; Wherein (a), (b) and (c) are as follows: (a) the distribution of the likelihood that any particular base pair in the human genome will appear at the end of the fragment to be sequenced; (b) the distribution of the likelihood that any pair of base pairs of the human genome will appear as the ends of a pair of sequenced fragments; (c) The distribution of the likelihood that any particular base pair in the human genome will appear in the sequenced fragment as a result of differentially occupied nucleosomes.

These and other embodiments are described in further detail below.

Figure 1 shows three types of information associating the cfDNA fragmentation pattern with nucleosomal occupancy, illustrated for the small genomic region. This same type of information may also occur through fragmentation of the chromatin with enzymes such as micrococose nuclease (MNase), DNase or transporter. Figure 1A shows the distribution of the likelihood that any particular base pair in the human genome will appear at the end of the fragment to be sequenced (i. E., The fragmentation point); Figure 1b shows the distribution of the likelihood that any pair of base pairs of the human genome will appear as a pair of ends of the fragment to be sequenced (i. E., A contiguous pair of fragmentation points producing individual molecules); Figure 1c shows the distribution of the likelihood (i. E., Relative coverage) that any particular base pair in the human genome will appear in a sequenced fragment as a result of different nucleosomal occupancy.
Figure 2 shows the insert size distribution of a typical cfDNA sequencing library, in which a pooled cfDNA sample (bulk.cfDNA) derived from human plasma containing an unknown number of healthy donors is shown.
Figure 3a shows the expression profiles of all cfDNA samples (plasma), cfDNA (tumor) from tumor patient samples, cfDNA (pregnancy) from pregnant female individuals, MNase (human cell line) and human DNA shotgun (shotgun) Shows the average periodogram intensity from Fast Fourier Transformation (FFT) of the read start coordinates mapped to the first (chrl) human autosomal chromosome over the Shotgun.
Fig. 3b shows the results of a cfDNA sample (plasma), cfDNA (tumor) from a tumor patient sample, cfDNA (pregnancy) from a pregnant female individual, MNase (human cell line) (FFT) of the read start coordinates mapped to the last (chr22) human autosomal region over the last (chr22) human autosomes.
Figure 4 shows the first three principal components (PC) of intensity in the periodicity of 196 base pairs (bp) in a 10 kilobase pair (kbp) block across all autosomes: Figure 4a shows PC 2 versus PC 1 And FIG. 4B shows PC 3 vs. PC 2.
Figure 5 shows a hierarchical clustering dendogram of the Euclidean distance of intensity measured at a cycle of 196 bp in a 10 kbp block across all autosomes.
Figure 6 shows the first three principal components of intensity at 18 < RTI ID = 0.0 > bp < / RTI > to 202 bp cycles in a 10 kbp block across all autosomes: Figure 6a shows PC 2 versus PC 1, Show.
Figure 7 shows a hierarchical clustering tangent of Euclidean distance of intensity measured at a period of 181 bp to 202 bp in a 10 kbp block across all autosomes.
Figure 8 shows the principal component analysis of intensities (first 7 out of 10 PCs) at 18 < RTI ID = 0.0 > bp < / RTI > to 202 bp periodicity in a 10 kbp block across all autosomal cfDNA datasets: Figure 8a shows PC 2 vs. PC 1 8B shows PC 3 vs. PC 2, FIG. 8C shows PC 4 vs. PC 3, FIG. 8D shows PC 5 vs. PC 4, FIG. 8E shows PC 6 vs. PC 5, FIG. 8F shows the PC 7 to the PC 6.
Figure 9 shows the principal component analysis of intensity at 181 bp to 202 bp periods in a 10 kbp block across all autosomal cells for the MNase data set: Figure 9a shows PC 2 vs. PC 1, Figure 9b shows PC 3 vs. PC 2, FIG. 9C shows PC 4 vs. PC 3, FIG. 9D shows PC 5 vs. PC 4, and FIG. 9E shows PC 6 vs. PC 5.
Figure 10 shows the average lumen intensity for a representative human autologous chromosome (chrll) across all synthetic cfDNA and MNase dataset mixtures.
Figure 11 shows the first two major components of intensity at 181 bp to 202 bp cycles in a 10 kbp block across all autosomes for a composite MNase dataset mixture.
Figure 12 shows the first two major components of intensity at 181 bp to 202 bp cycles in a 10 kbp block across all autosomes for a synthetic cfDNA dataset mixture.
Figure 13 shows a hierarchical clustering dendogram of Euclidean distances of intensity at 181 bp to 202 bp cycles in a 10 kbp block across all autosomes for the synthetic MNase and cfDNA mixture data sets.
Figure 14 shows the read starting density at 1 kbp window around 23,666 CTCF binding sites for a sample set having at least 100M readings.
Figure 15 shows the read starting density at 1 kbp window around 5,644 c-Jun binding sites for a sample set with at least 100M readings.
Figure 16 shows the read initiation density for a 1 kbp window around 4,417 NF-YB binding sites for a set of samples with a reading of at least 100M.
Figure 17 shows a schematic outline of the process of generating cfDNA fragments. Apoptosis and / or necrotic cell death results in almost complete digestion of the native chromatin. In general, protein-bound DNA fragments that associate with histones or transcription factors preferentially survive digestion and release into the circulatory system during the disappearance of naked DNA. The fragments may be recovered from peripheral blood plasma following proteinase treatment. In healthy individuals, cfDNA is primarily derived from the bone marrow and lymphoid cell lines, but may be derived from one or more additional tissues in certain medical conditions.
Figure 18 shows the fragment length of cfDNA observed with conventional sequencing library preparation. The length is deduced in the alignment of the paired end sequenced readings. At 167 base pairs (bp) (green dotted line), the reproducible peak of the fragment length is consistent with the binding with the chromatogram. An additional peak demonstrates a ~ 10.4 bp cycle corresponding to the helical pitch of DNA on the nucleosome core. End-repair of the enzyme during library preparation can eliminate the 5 ' and 3 ' overhangs and damage the actual cleavage site.
Figure 19 shows the 167 bp fragment in a conventional library and the dinucleotide composition of an adjacent genomic sequence. The frequency of dinucleotides observed in the BH01 library was compared to the expected frequency from the simulated fragment (matching for endpoint biases due to both cleavage and adapter ligation preferences).
Figure 20 shows a schematic of a single strand library manufacturing protocol for cfDNA fragments.
Figure 21 shows fragment lengths of cfDNA observed in the preparation of single-stranded sequencing libraries. Enzymatic terminal repair of the template molecule during library preparation is not performed. Short fragments of 50-120 bp are much more abundant than conventional libraries. While the periodicity of ~ 10.4 bp is maintained, its phase shifts by ~ 3 bp.
Figure 22 shows the 167 bp fragment in a single-stranded library and the dinucleotide composition of an adjacent genomic sequence. The observed frequency of dinucleotides in the IH02 library was again compared to the expected frequency derived from the matched simulated fragment for endpoint bias. The apparent difference in the background level of the bias between BH01 and IH02 is related to the difference between the simulations rather than the actual library (not data).
Figure 23A shows a gel image of a representative cfDNA sequencing library made with a conventional protocol.
Figure 23B shows a gel image of a representative cfDNA sequencing library made with a single-stranded protocol.
24A shows the mononucleotide cleavage of the cf DNA fragment.
Figure 24b shows the dinucleotide cleavage bias of the cfDNA fragment.
Figure 25 shows a schematic outline of the estimation of the nucleosome arrangement. The windowed protection score per base (WPS) is calculated by subtracting the number of fragment endpoints within the 120 bp window from the number of fragments that completely span the window. High WPS values indicate an increase in DNA protection from digestion; low values indicate that the DNA is not protected. The peak call identifies the adjacent region of the raised WPS.
Figure 26 shows nucleosomes strongly positioned in a well studied alpha-satellite array. The application range of sample CH01, the fragment endpoint, and the length of the fragment CHO for a short fragment (120 bp window, 120-180 bp reading) or a short fragment (16 bp window, 35-80 bp reading) bin in the locus surrounding the chromosome on chromosome 12 The WPS value is displayed. The nucleosome call (middle, blue box) from CH01 is regularly spaced across the locus. Nucleosome calls based on MNase digestion from two published studies (middle, purple and black boxes) are also displayed. This locus overlaps with the annotated alpha-satellite array.
Figure 27 shows the putative nucleosome arrangement located around the DNase I hypersensitive site (DHS) on chromosome 9. The coverage of the sample CH01, the fragment endpoint, and the WPS value are displayed for long and short fragment beans. The gray highlighted areas of sensitivity are indicated by reduced coverage in long fragment beans. The nucleosome call of CH01 (middle, blue box) adjacent to DHS is more widely spaced than a typical adjacent pair, consistent with the accessibility of intervening sequences to regulatory proteins including transcription factors. The coverage of shorter fragments that may be associated with these proteins is increased in DHS, which overlaps with some annotated transcription factor binding sites (not shown). From the two published studies, the nucleosome call based on MNase digestion is as shown in Fig.
28 shows a schematic diagram of peak calling and scoring according to one embodiment of the present disclosure;
Figure 29 shows CH01 peak density by GC content.
Figure 30 shows a histogram of the distance between adjacent peaks by the sample. The distance is measured from the peak call to the adjacent call.
Figure 31 shows a comparison of peak calls between samples. For each sample pair, the distance between each peak call of the sample with the smallest number of peaks and the closest peak call of the other sample is calculated and visualized with a histogram of bin size 1. Negative numbers indicate that the nearest peak is upstream, and positive numbers indicate that the nearest peak is downstream.
Figure 32 shows a comparison of peak calls between samples. FIG. 32A shows IH01 versus BH01, FIG. 32B shows IH02 versus BH01, and FIG. 32C shows IH02 versus IH01.
Figure 33A shows the nucleosome score for an actual vs. simulated peak.
33B shows the center peak offset (left y-axis) in the score bin as a function of the score bin and the number of peaks in each score bin (right y-axis).
34 shows a comparison of the peak calls between the sample and the matched simulation, FIG. 34A shows the BH01 simulation versus BH01 realization, FIG. 34B shows the IH01 simulation versus IH01 realization and FIG. 34C shows the IH02 simulation versus IH01 real Lt; / RTI >
Figure 35 shows the distance between adjacent peaks, sample CH01. The dotted black line shows the distribution mode (185bp).
Figure 36 shows an aggregated and adjusted window protection score (WPS; 120 bp window) around 22,626 transcriptional start sites (TSS). After adjusting the strand and transfer direction, the TSS is aligned to the 0 position. The aggregated WPS is tabulated for both actual and simulated data, summing the WPS per TSS at each location relative to the central TSS. The floated value represents the difference between the actual and simulated aggregated WPS and is further adjusted to a local background as described in more detail below. Higher WPS values show preferential protection from disconnection.
Figure 37 shows the adjusted WPS aggregated around 22,626 start codons.
Figure 38 shows the aggregated adjusted WPS around 224,910 splice donor sites.
Figure 39 shows the aggregated adjusted WPS around 224,910 splice acceptor sites.
Figure 40 shows adjusted WPS aggregated for various gene characteristics using data from CH01, including actual data, matched simulations, and the differences.
Figure 41 shows the nucleosome spacing in the A / B section. The median value of the nucleosome spacing in a non-overlapping 100 kilobase (kb) bean (each containing ~ 500 nucleosome calls) is calculated over the entire genome. The A / B block predictions for GM12878 with 100 kb resolution are presented from published data. Compartment A is associated with an open chromatin, and compartment B is associated with a closed chromatin.
Figure 42 shows the nucleosome spacing and A / B compartments on chromosome 7 and 11. The A / B segmentation (red and blue bars) is a summary of mainly chromosome G-banding (ideograms, gray bars). The center nucleosome interval (black dot) is calculated from the 100 kb bin and plotted over the A / B segmentation.
Figure 43 shows the aggregated adjusted WPS for the 93,550 CTCF sites in the long (upper) and short (bottom) fractions.
Figure 44 shows an enlarged view of the aggregated tailored WPS for the short cut cfDNA at the CTCF site. A bright red bar (and corresponding shade in the plot) represents the location of a known 52 bp CTCF binding motif. The dark red sub-portion of this bar represents the position of the 17 bp motif used in the FIMO motif search.
Figure 45 overlaps ENCODE ChIP-seq peak (93,530 sites) -1 to +1 nucleosome spacing calculated around the CTCF site derived from the clustered FIMO predicted CTCF site (purely motif-based: 518,632 sites) (23,723 sites) that were experimentally observed to be active across the 19 cell lines. The least stringent set of CTCF sites is usually separated by approximately the same distance as the entire genome (~ 190 bp). However, under the most stringent conditions, most of the CTCF sites are separated by a much wider range (~260 bp), consistent with the active CTCF binding and the positional change of the adjacent nucleosomes.
Figures 46-48 show the CTCF occupancy location adjacent to the nucleosomes: Figure 46 shows the peak-to-peak distance for the three closest upstream and the three closest downstream peak calls for 518,632 CTCF binding sites predicted by FIMO Lt; / RTI > Figure 47 shows the peak-to-peak distances for the three closest upstream and three closest downstream peak calls for 518,632 CTCF binding sites predicted by FIMO, as in Figure 46, where the same set of CTCF sites are 93,530 And filtered based on overlap with ENCODE ChIP-seq peak. Figure 48 shows the peak-to-peak distances for three nearest upstream and three nearest downstream peak calls for 93,530 CTCF binding sites predicted by FIMO, as in Figure 47, where the CTCF site set was divided into 19 cell lines Lt; RTI ID = 0.0 > CTCF < / RTI >
Figure 49 shows that for a subset of the predicted CTCF regions with adjacent spacing (230-270 bp) of the adjacent nucleosomes, both the long (top) and short (bottom) fractions are located in a more rigid subset of the CTCF region Indicating a stronger placement signal. See FIG. 45 for a key that defines a color line.
Figures 50-52 show the CTCF occupancy location adjacent to the nucleosomes: Figure 50 shows the average short fraction WPS (upper panel) and the mean long fraction WPS (lower panel) for 518,632 sites, And a distance bin representing the number of base pairs separating adjacent +1 and -1 nucleosome calls. Figure 51 shows the average short fraction WPS (upper panel) and the mean long fraction WPS (lower panel) for 518,632 sites in Figure 50, where the same set of CTCF sites are based on overlap with the ENCODE ChIP-seq peak Filtered. Figure 52 shows the mean short fraction WPS (upper panel) and the mean long fraction WPS (lower panel) for the regions of Figure 51, where the same set of sites was observed for the active CTCF sites experimentally observed over 19 cell lines Lt; RTI ID = 0.0 > set. ≪ / RTI > The key for defining the color line in Fig. 50 is the same in Fig. 51 and Fig.
Figures 53a-h show the footprint of the transcription factor binding site from short and long cfDNA fragments. Clustered FIMO binding site predictions intersected the ENCODE ChIP-seq data to obtain a reliable set of transcription factor (TF) binding sites for a set of additional factors. The pooled adjusted WPS for the region adjacent to the set of generated TF binding sites is displayed for both long and short fragments of the cfDNA fragment. The higher WPS values each represent a higher probability of occupying the nucleosome or TF. 53A: AP-2; Figure 53b: E2F-2; Figure 53c: EBOX-TF; 53D: IRF; 53E: MYC-MAX; 53F: PAX5-2; Figure 53g: RUNX-AML; 53H: YY1.
Figure 54 shows the aggregated adjusted WPS for the transcription factor ETS (210,798 sites). The calculated WPS from both the long (top) and short (bottom) cfDNA fractions is displayed. Signals consistent with TF protection in the tissues of the surrounding nucleosomes (long fractions) and the binding site itself (short fractions) are observed. A similar analysis for additional TF is shown in Figures 53a-h.
Figure 55 shows the aggregated adjusted WPS for the transcription factor MAFK (32,159 sites). The calculated WPS from both the long (top) and short (bottom) cfDNA fractions is displayed. Signals consistent with TF protection in the tissues of the surrounding nucleosomes (long fractions) and the binding site itself (short fractions) are observed. A similar analysis for additional TF is shown in Figures 53a-h.
Figure 56 shows an estimate of a mixture of cell types that contribute to acellular DNA based on the DNase hypersensitive (DHS) site. The frequency distribution of the peak-to-peak spacing of the nucleosome calls at the DHS site from the 116 diverse biological samples exhibits a bimodal distribution and the second mode is the nucleosome spread at the active DHS site due to intervening transcription factor binding (~ 190 bp to 260 bp). The DHS region identified in lymphocytes or bone marrow samples has the largest proportion of DHS sites with a broad nucleosome gap consistent with hematopoietic cell death as a major source of cfDNA in healthy individuals.
Figure 57 shows that the segmentation of the adjusted WPS score around the transcription initiation site (TSS) to the five gene expression bins (fifth quintile) defined for NB-4 (acute promyelocytic leukemia cell line) To show the difference between the two. Highly expressed genes show a strong phase of nucleosomes in the transcript. Upstream of TSS, the -1 nucleosome is well located throughout the expression bell, while the -2 and -3 nucleosomes are well located only in the medium to highly expressed genes.
FIG. 58 shows that, for intermediate to highly expressed genes, a short cutoff peak was observed between TSS and the -1 nucleosome in agreement with the footprint of the transcriptional pre-initiation complex or some of its components in the transcriptionally active gene.
Figure 59 shows that the median nucleosomal distance in the transcript is negatively correlated with the gene expression measured for the NB-4 cell line (p = -0.17, n = 19,677 genes). Almost to no expression of the gene shows a neutralization distance of 193 bp, while for the expressed gene this distance is 186-193 bp. This negative correlation is particularly important when more nucleosome calls are used (e.g., at least 60 nucleosomes are needed, p = -0.50, n = 12,344 genes ), And stronger.
Figure 60 illustrates a fast Fourier transform (FFT) to quantify the specific frequency contribution (intensity) in a long fragment WPS for the first 10 kb of the gene body starting at each TSS to deconvolate a plurality of contributions. (FFT). ≪ / RTI > The trajectory for the correlation between RNA expression and primary tissue with different frequencies of this intensity in 76 cell lines is presented. The dark black line indicates the NB-4 cell line. The correlation is strongest in the 193-199 bp frequency range.
Figure 61 shows estimates of cell types contributing to cell-free DNA in healthy conditions and cancer. The top panel displays the ranking of the correlations for the 76 RNA expression data sets with average intensity in the 193-199 bp frequency range for the various cfDNA libraries sorted by type and listed from top (top row) to lowest (bottom row) Lt; / RTI > The correlation values and the overall cell line or tissue name are given in Table 3. The strongest correlation to all three healthy samples (BH01, IH01 and IH02; the first three columns) is associated with lymphocytes and myeloid cell lines as well as bone marrow. In contrast, cfDNA samples (IC15, IC17, IC20, IC35, IC37; the last five columns) obtained from patients with stage IV cancer show the highest correlation with various cancer cell lines, for example IC17 (hepatocellular carcinoma, HCC) HepG2 (hepatocellular carcinoma cell line) and IC35 (ductal carcinoma, DC) have the highest correlation with MCF7 (metastatic breast adenocarcinoma cell line). When comparing the cell line / tissue rankings observed in the cancer samples to each of the three healthy samples and averaging the ranking changes (bottom panel), the highest ranking change was observed by comparing the three healthy samples to each other, Which is more than twice as high as that of ' For example, in the case of IC15 (small cell lung carcinoma, SCLC), the rank of SCLC-21H (small cell lung carcinoma cell line) increased by an average of 31 positions, and in case of IC20 (squamous cell lung carcinoma, SCC) (Metastatic breast adenocarcinoma cell line) increased by 21 in average, and HepG2 increased by 24 in IC37 (colorectal adenocarcinoma, AC).
Figure 62 shows the quantification of aneuploidy for selecting samples with high burden of circulating tumor DNA, based on coverage (Figure 62a) or allele balance (Figure 62b). 62A is a graph showing the relationship between the number of observed readings for each sample (black dot) compared to the simulated sample (red dot) assumed to be insoluble, and Z Show the sum of the scores. Figure 62B shows the allele balance at each of the 48,800 common SNPs evaluated per chromosome for a subset of the samples selected for further sequencing.
Figure 63 shows a comparison of peak calls for the published set of nucleosome calls. 63A is a graphical representation of three published data sets ([Gaffney et al., 2012], [JS Pedersen et al., 2014] and [A Schep et al. 2015]) of the nucleosome peak. Previously published data sets probably do not represent a single defined mode at ~ 185 bp nucleosomal distance from the regular because of poor sampling or wide call coverage. By contrast, all nucleosome calls in cfDNA show one well-defined mode. The matched simulated data set has a shorter mode (166 bp) and wider distribution. In addition, the higher the coverage of the cfDNA data set used to generate the call, the higher the percentage of calls displayed in the distribution mode. Figure 63b shows the number of nucleosomes for each set list identical to Figure 63a. The cfDNA nucleosome callo presents the most comprehensive call set with a nearly 13 M nucleosome peak call. Figure 63c shows the distance between each peak call in the IH01 cfDNA sample and the closest peak call from the three previously published data sets. 63D shows the distance between each peak call in the IH02 cfDNA sample and the closest peak call from the three previously published data sets. 63E shows the distance between each peak call in the BH01 cfDNA sample and the closest peak call from the three previously published data sets. Figure 63f shows the distance between each peak call in the CH01 cfDNA sample and the closest peak call from the three previously published data sets. Figure 63g shows the distance between each peak call in the CA01 cfDNA sample and the closest peak call from the three previously published data sets. A negative number indicates that the nearest peak is upstream, and a positive number indicates that the nearest peak is downstream. As cfDNA coverage increases, a higher proportion of previously published calls appears closer to the determined nucleosome call. The best match can be found in Gaffney et al., PLoS Genet., Vol. 8, e1003036 (2012) and A Schep et al. (2015)]. Figure 63h shows the distance between each peak call and the closest peak call from the three previously published data sets, but this time the distance to CA01's matched simulation. The closest actual nucleosomal location can be found in Gaffney et al., PLoS Genet., Vol. 8, e1003036 (2012)] and JS Pedersen et al., Genome Research, vol. 24, pp. 454-466 (2014)] tend to move away from the peak in the simulation for the call. The calls generated by A Schep et al. (2015) show some overlap with the simulated call.

DETAILED DESCRIPTION OF THE INVENTION

The present disclosure provides methods for determining one or more tissues and / or cell types that produce acellular DNA in a biological sample of a subject. In some embodiments, the disclosure provides a method for identifying a disease or disorder in a subject as a function of one or more determined tissues and / or cell types associated with cfDNA in a biological sample from the subject.

This disclosure is based on the prediction that the cf DNA molecules from different cell types or tissues are different for (a) the distribution of the likelihood that any particular base pair within the human genome will appear at the end (fragmentation point) of the cfDNA fragment; (b) the distribution of the likelihood that any pair of base pairs of the human genome will appear at the end of one of the cfDNA fragments (i. e., a pair of successive fragmentation points producing individual cfDNA molecules); (c) The distribution of the likelihood that any particular base pair in the human genome will appear in a cfDNA fragment (i. e., relative coverage) as a result of a difference in nucleosome share. These are referred to below as distributions (a), (b) and (c), or collectively referred to as "nucleosome-dependent cleavage probability map", "cleavage accessibility map" or "nucleosome map" 1). For reference, the nucleosome map is determined through sequencing of fragments derived from fragmentation of the chromatin using an enzyme such as micrococose nuclease (MNase), DNase, or transposon, Can be measured by an equivalent procedure that preferentially fragment genomic DNA at or between the boundaries of the chromatosome.

In healthy individuals, cfDNA stems predominantly from the apoptosis of blood cells, hematopoietic lineage cells. When these cells undergo apoptosis programmed, their genomic DNA is cleaved, released into the circulatory system, and subsequently degraded by the nuclease. The length distribution of cfDNA oscillates at a cycle of about 10.5 base pairs (bp) corresponding to the helical pitch of the DNA wrapped around the nucleosome and around 167 bp corresponding to the length of DNA associated with the linker- associated mononucleosomes (Fig. 2). This evidence has led to the hypothesis that association of cfDNA with nucleosomes protects cfDNA from complete and rapid degradation in the circulatory system. Another possibility is that the length distribution simply arises from the DNA cleavage pattern during apoptosis itself, which is directly affected by the nucleosome arrangement. In any case, the length distribution of cfDNA provides clear evidence that the fragmentation process that produces cfDNA is affected by the nucleosome arrangement.

In some embodiments, the disclosure is directed to a method of fragmenting a chromatin with an enzyme such as cfDNA or micrococose nuclease (MNase), DNase, or transposase derived from library construction and body fluids, or by cleaving the chromosome (A), (b) and / or (c) by massive parallel sequencing of DNA derived from equivalent procedures that preferentially fragment genomic DNA in between. As described below, such a distribution may include, for example, a transcription factor binding site, a gene model feature (e.g., a transcription initiation site or gene body), a topologically associated domain Collecting or summarizing periodic signals of nucleosomal locations within various subsets of the genome, such as quantifying the periodicity in a discontinuous subset of genomes defined by tissue expression data or other correlates of nucleosome positioning Can be "transformed" It can also be defined by tissue specific data. For example, signals can be collected or summarized near tissue-specific DNase I hypersensitive sites.

This disclosure provides a detailed genome-wide map of in vivo nucleosome protection as deduced from plasma-containing cfDNA fragments. The CHO1 map derived from the cfDNA of healthy individuals contains a uniformly spaced local maximum of about 13 M of nucleosome protection over the majority of the mapped human reference genome. The number of peaks was essentially saturated in CH01, but other quality criteria continued as a function of sequence depth (Figs. 33a-b). Thus, in the same way, based on almost all of the cfDNA sequence determinations we have conducted so far for this and other studies ('CA01', 14.5 billion (G) fragments; 700-fold coverage, 13.0 M peaks) A further nucleosome map was constructed across the entire genome. (Figs. 33a-b, 63a-h), the present inventors have found that this is based on cfDNA from both healthy and non-healthy individuals, although this map shows a much more uniform spacing and higher supported peak call (Table 1 and 5).

A dense genome-wide map of the nucleosome protection disclosed herein is intended to provide a more comprehensive approach to the saturation of the mappable portion of the human reference genome, and to a greater degree than previous efforts to map the entire human genome of nucleosome placement or protection (Fig. 63a-h) that are substantially uniform and consistent with predicted nucleosome repeat lengths. In contrast to nearly all previous efforts, the fragments observed here are produced by an endogenous physiological process, and thus are less likely to be applied to technical variations associated with in vitro microcococcus nuclease digestion. Cell types that cause cfDNA to be considered in this reference map are necessarily heterogeneous (for example, a mixture of lymphocytes and myeloid cell types of healthy individuals). Nonetheless, the relative completeness of the maps can facilitate a deeper understanding of the nucleosome arrangement and spacing in human cells, and the process of directing the interaction of nucleosomes with nucleosomal structures and transgenic genetic control have.

Object  In biological samples cfDNA  How to determine the source (s)

As generally discussed above, and as more specifically demonstrated in the Examples below, the techniques of the present invention can be used to determine (or identify) the tissue (s) and / or cell type (s) that contribute to cfDNA in a biological sample of a subject For example, prediction).

Thus, in some embodiments, this disclosure provides a method of determining the tissue and / or cell type that produces acellular DNA (cfDNA) in a subject, comprising isolating cfDNA from a biological sample from the subject, Wherein the isolated cfDNA comprises a plurality of cfDNA fragments; Determining a sequence associated with at least a portion of a plurality of cfDNA fragments; determining the genomic location in the reference genome for at least a portion of the cfDNA fragment endpoints of the plurality of cfDNA fragments as a function of the cfDNA fragment sequence; And determining at least a portion of the tissue and / or cell type that produces the cfDNA fragment as a function of the genomic location of at least a portion of the cfDNA fragment endpoint.

In some embodiments, the biological sample comprises, consists essentially of, or consists of whole blood, peripheral blood plasma, urine or cerebrospinal fluid.

In some embodiments, determining at least a portion of the tissue and / or cell type that produces the cfDNA fragment comprises comparing the mathematical transformation of the genomic location of at least a portion of the cfDNA fragment endpoint, or its distribution, with one or more reference maps do. As used herein, the term " reference map "refers to any type or form of a cfDNA sequence that can be correlated or compared with the characteristics of cfDNA in a biological sample of a subject as a function of coordinates in the genome (e.g., the reference genome) Data. The reference map can be correlated or compared with the characteristics of cfDNA in the biological sample of the subject by any suitable means. By way of non-limiting example, a correlation or comparison may be performed either directly or after performing a mathematical transformation on its distribution across windows in the reference genome, followed by a numerical or reference map to any other Can be achieved by analyzing the frequency of the cfDNA endpoints in the biological sample of the subject in terms of the status. In another non-limiting example, the correlation or comparison is based on the nucleosome interval (s) determined based on the cfDNA of the biological sample of the subject in light of the determined nucleosome interval (s), or the nucleosome interval And other characteristics that are correlated with each other.

The reference map (s) may include, for example, an open database of genomic information, any published data, or any (including, but not limited to) data generated for a particular population of reference objects And may be supplied or derived from a suitable data source. In some embodiments, the reference map comprises a DNase I sensitive data set. In some embodiments, the reference map comprises an RNA expression data set. In some embodiments, the reference map includes a chromosome stereochemical map. In some embodiments, the reference map includes a chromatic accessibility map. In some embodiments, the reference map comprises data generated from at least one tissue or cell type associated with the disease or disorder. In some embodiments, the reference map comprises the location of a nucleosome and / or a chromatome in a tissue or cell type. In some embodiments, the reference map is generated by a procedure involving digesting the chromatin with an exogenous nuclease (e. G., Micrococus nuclease). In some embodiments, the reference map includes chromatin accessibility data determined by a transposition-based method (e.g., ATAC-seq). In some embodiments, the reference map comprises data associated with DNA binding to the tissue or cell type and / or the location of the DNA occupancy protein. In some embodiments, the DNA binding and / or the DNA occupying protein is a transcription factor. In some embodiments, the site is determined by a procedure involving chromatin immunoprecipitation of the cross-linked DNA-protein complex. In some embodiments, the site is determined by a procedure involving treating DNA associated with the tissue or cell type with a nuclease (e. G., DNase-I). In some embodiments, the reference map is created by sequencing cf DNA fragments from a biological sample from one or more individuals with known disease. In some embodiments, the biological sample from which a reference map is drawn is collected from a human cell or tissue xenografted animal.

In some embodiments, the reference map comprises biological features corresponding to the location of a DNA binding or DNA occupancy protein for a tissue or cell type. In some embodiments, the reference map comprises biological features corresponding to quantitative RNA expression of one or more genes. In some embodiments, the reference map includes biological features corresponding to the presence or absence of one or more histone marks. In some embodiments, the reference map comprises biological features corresponding to hypersensitivity to nuclease cleavage.

The step of comparing the genomic location of at least a portion of the cfDNA fragment end point with one or more reference maps can be performed in a variety of ways. In some embodiments, the cfDNA data generated from the biological sample (e.g., the genomic location of the cfDNA fragment, the endpoint thereof, the frequency of its endpoint and / or the nucleosome interval (s) estimated from its distribution) It is compared to a reference map. In this embodiment, the tissue or cell type associated with the reference map having the highest correlation with the cfDNA data in the biological sample is considered to contribute. By way of non-limiting example, if the cfDNA data comprises a similar cfDNA endpoint list and a position within its reference genome, the reference map (s) having the most similar list of positions in the cfDNA endpoint and its reference genome can be considered to contribute have. As another non-limiting example, reference map (s) having the highest correlation (or increased correlation relative to cfDNA from a healthy subject) with the mathematical transformation of the distribution of cfDNA fragment endpoints from a biological sample is contributed Can be considered. The tissue type and / or cell type corresponding to the reference map is considered to be a potential source of cfDNA isolated in the biological sample.

In some embodiments, determining at least a portion of the tissue and / or cell type that produces the cfDNA fragment comprises performing a mathematical transformation on the distribution of the genomic location of at least a portion of the cfDNA fragment endpoint. One non-limiting example of a mathematical transformation suitable for use in connection with the teachings of the present invention is a Fourier transform, such as a Fast Fourier Transform ("FFT").

In some embodiments, the method further comprises determining a score for each of the coordinates of at least a portion of the reference genome, wherein the score is determined as a function of at least a plurality of cfDNA fragment endpoints and their genomic locations, The step of determining at least a portion of the tissue and / or cell type that produces the < RTI ID = 0.0 > cfDNA < / RTI > fragment comprises comparing the score with one or more reference maps. The score may be any criteria (e.g., a numerical ranking or probability) that can be used to assign a relative or absolute value to the coordinates of the reference genome. For example, the score may be a probability that the coordinates represent the location of the cfDNA fragment endpoint, or that the coordinates represent the location of the genome that is preferentially protected from the nuclease cleavage by nucleosome or protein binding, Lt; / RTI > As another example, a score may be associated with a nucleosome gap in the particular region, as determined by mathematical transformation of the cfDNA fragment endpoint distribution within a particular region of the genome. Such a score may be determined by any suitable method, including counting absolute or relative events (e.g., the number of cfDNA fragment endpoints) associated with that particular coordinate, or performing a mathematical transformation on the value of the coefficient in the region or genomic coordinates Can be assigned to the coordinates. In some embodiments, the score for the coordinates is related to the probability that the coordinates are the positions of the cfDNA fragment endpoints. In another embodiment, the score for the coordinates is related to the probability that the coordinates represent the location of the genome that is preferentially protected from nuclease cleavage by nucleosome or protein binding. In some embodiments, the score is related to the nucleosome spacing in the genome region of the coordinates.

The tissue (s) and / or cell type (s) mentioned in the methods described herein may be any tissue or cell type that produces cfDNA. In some embodiments, the tissue or cell type is a primary tissue from a subject having a disease or disorder. In some embodiments, the disease or disorder is selected from the group consisting of cancer, normal pregnancy, complications of pregnancy (e. G., Ischemic pregnancy), myocardial infarction, inflammatory bowel disease, systemic autoimmune disease, local autoimmune disease, Allograft, < RTI ID = 0.0 > stroke, < / RTI > and localized tissue injury without rejection.

In some embodiments, the tissue or cell type is a primary tissue from a healthy subject.

In some embodiments, the tissue or cell type is an immortalized cell line.

In some embodiments, the tissue or cell type is a biopsy from a tumor.

In some embodiments, the reference map is based on sequence data obtained from a sample obtained from at least one reference object. In some embodiments, the sequence data defines the location of the cfDNA fragment endpoint in the reference genome, for example, when a reference map is generated by sequencing cfDNA from the subject (s) with a known disease. In another embodiment, the sequence data on which the reference map is based is generated by chromatin digestion of a DNase I hypersensitive site data set, an RNA expression data set, a chromosome stereochemistry map, or a chromosome accessibility map, or a micrococycle nuclease And a nucleosome mapping map created.

In some embodiments, the reference object is healthy. In some embodiments, the reference subject is selected from the group consisting of an allogeneic transplant involving cancer, normal pregnancy, complication of pregnancy (e. G., Isomeric pregnancy), myocardial infarction, inflammatory bowel disease, systemic autoimmune disease, local autoimmune disease, , Allografts that do not involve rejection, stroke and localized tissue injury.

In some embodiments, the reference map includes a score for at least a portion of the coordinates of the reference genome associated with the tissue or cell type. In some embodiments, the reference map includes a mathematical transformation of the score, such as a Fourier transform of the score. In some embodiments, the score is based on annotations of reference genomic coordinates for tissue or cell type. In some embodiments, the score is based on the location of the nucleosome and / or the chromatome. In some embodiments, the score is based on a transcription initiation site and / or a transcription termination site. In some embodiments, the score is based on the predicted binding site of at least one transcription factor. In some embodiments, the score is based on predicted nuclease hypersensitive sites. In some embodiments, the score is based on the predicted nucleosome spacing.

In some embodiments, the score is associated with at least one orthogonal biological characteristic. In some embodiments, orthogonal biological features are associated with highly expressed genes. In some embodiments, orthogonal biological features are associated with a low expression gene.

In some embodiments, at least some of the plurality of scores have a value in excess of the threshold (minimum) value. In such an embodiment, the score below the threshold (min) value is excluded from comparing the score to the reference map. In some embodiments, the threshold value is determined prior to determining the tissue (s) and / or cell type (s) that produce the cfDNA. In another embodiment, the threshold value is determined after determining the tissue (s) and / or cell type (s) that produce cfDNA.

In some embodiments, determining the tissue and / or cell type that produces cfDNA as a function of the plurality of genomic locations of at least a portion of the cfDNA fragment endpoint comprises determining a mathematical transformation of the genomic location distribution of at least a portion of the cfDNA fragment endpoint of the sample With one or more features of the one or more reference maps. One non-limiting example of a mathematical transformation suitable for this purpose is a Fourier transform, such as a Fast Fourier Transform ("FFT").

In any of the embodiments described herein, the method may further comprise generating a report comprising a list of determined tissues and / or cell types that produce the isolated cfDNA. The report may optionally include a sample and / or a tissue, a tissue that is likely not to produce any cfDNA isolated from the biological sample, the type of biological sample, the date the biological sample was obtained from the subject, the date the cfDNA isolation step was performed and / (S) and / or any other information about the cell type (s).

In some embodiments, the reports include, but are not limited to, proposals for additional diagnostic tests, proposals for initiation of treatment regimens, suggestions for alteration of existing therapies and / or proposals to delay or discontinue existing therapies, Additional recommended treatment protocols are included.

From the object  How to Identify a Disease or Disorder

As generally discussed above and more specifically demonstrated in the Examples below, the techniques of the present invention are based, at least in part, on the type of tissue and / or cell that contribute to cfDNA in a biological sample of a subject, , Or to determine (e.g., predict) the absence of a disease or disorder.

Thus, in some embodiments, the disclosure provides a method of identifying a disease or disorder in a subject, the method comprising isolating cell-free DNA (cfDNA) from a biological sample from the subject, wherein the isolated cfDNA comprises a plurality of cfDNA Fragments; Determining a sequence associated with at least a portion of a plurality of cfDNA fragments; determining the genomic location in the reference genome for at least a portion of the cfDNA fragment endpoints of the plurality of cfDNA fragments as a function of the cfDNA fragment sequence; determining at least a portion of the tissue and / or cell type that produces cfDNA as a function of genomic location of at least a portion of the cfDNA fragment endpoint; And identifying the disease or disorder as a function of the determined tissue and / or cell type producing the cfDNA.

In some embodiments, the biological sample comprises, consists essentially of, or consists of whole blood, peripheral blood plasma, urine or cerebrospinal fluid.

In some embodiments, determining the tissue and / or cell type that produces the cfDNA comprises comparing the mathematical transformation of the genomic location of at least a portion of the cfDNA fragment endpoint or its distribution to one or more reference maps. The term "reference map " used in connection with these embodiments may have the same meaning as described above in connection with a method for determining tissue (s) and / or cell type (s) that produce cfDNA in a biological sample of a subject have. In some embodiments, the reference map includes at least one of a DNase I hypersensitive site data set, an RNA expression data set, a chromosome stereotactic map, a chromosome accessibility map, sequence data generated from a sample from at least one reference object, Enzyme-mediated fragmentation data corresponding to one tissue, and / or the location of a nucleosome and / or chromatome within a tissue or cell type. In some embodiments, the reference map is created by sequencing cf DNA fragments from a biological sample from one or more individuals with known disease. In some embodiments, the biological sample from which a reference map is drawn is collected from a human cell or tissue-implanted animal.

In some embodiments, the reference map is generated by digesting the chromatin with an exogenous nuclease (e. G., Micrococus nuclease). In some embodiments, the reference map includes chromatin accessibility data determined by a displacement-based method (e.g., ATAC-seq). In some embodiments, the reference map comprises data associated with DNA binding to the tissue or cell type and / or the location of the DNA occupancy protein. In some embodiments, the DNA binding and / or DNA occupying protein is a transcription factor. In some embodiments, the site is determined by chromatin immunoprecipitation of the cross-linked DNA-protein complex. In some embodiments, the site is determined by treating the DNA associated with the tissue or cell type with a nuclease (e. G., DNase-I).

In some embodiments, the reference map comprises biological features corresponding to the location of a DNA binding or DNA occupancy protein for a tissue or cell type. In some embodiments, the reference map comprises biological features corresponding to quantitative expression of one or more genes. In some embodiments, the reference map includes biological features corresponding to the presence or absence of one or more histone marks. In some embodiments, the reference map comprises biological features corresponding to hypersensitivity to nuclease cleavage.

In some embodiments, determining the tissue and / or cell type that produces cfDNA comprises performing a mathematical transformation on the distribution of the genomic location of at least a portion of the plurality of cfDNA fragment endpoints. In some embodiments, the mathematical transform includes a Fourier transform.

In some embodiments, the method further comprises determining a score for each of at least some of the coordinates of the reference genome, wherein the score is determined as a function of at least a plurality of cfDNA fragment endpoints and their genomic locations, Determining at least a portion of the tissue and / or cell type that produces the cfDNA fragment comprises comparing the score with one or more reference maps. The score may be any criteria (e.g., a numerical ranking or probability) that can be used to assign a relative or absolute value to the coordinates of the reference genome. For example, the score may consist of a probability that the coordinates represent the position of the cfDNA fragment endpoint, or the probability that the coordinates represent the position of the genome that is preferentially protected from the nuclease cleavage by the nucleosome or protein binding, ≪ / RTI > As another example, the score may be related to the nucleosome spacing in that particular region, as determined by mathematical transformation of the cfDNA fragment endpoint distribution within a particular region of the genome. Such a score may include, for example, any number of absolute or relative events (e. G., The number of cfDNA fragment endpoints) associated with the particular coordinate, or any mathematical transformation involving performing a mathematical transformation on the value of the coefficient Can be assigned to the coordinates by a suitable method. In some embodiments, the score for the coordinates is related to the probability that the coordinates are the positions of the cfDNA fragment endpoints. In another embodiment, the score for the coordinates is related to the probability that the coordinates represent the location of the genome that is preferentially protected from nuclease cleavage by nucleosome or protein binding. In some embodiments, the score is related to the nucleosome spacing in the genome region of the coordinates.

The term "score" used in connection with these embodiments may have the same meaning as described above in connection with the method of determining tissue (s) and / or cell type (s) producing cfDNA in a biological sample of a subject have. In some embodiments, the score for the coordinates is related to the probability that the coordinates are the positions of the cfDNA fragment endpoints. In another embodiment, the score for the coordinates is related to the probability that the coordinates represent the location of the genome that is preferentially protected from nuclease cleavage by nucleosome or protein binding. In some embodiments, the score is related to the nucleosome spacing in the genome region of the coordinates.

In some embodiments, the tissue or cell type used to generate the reference map is a primary tissue from a subject having a disease or disorder. In some embodiments, the disease or disorder is selected from the group consisting of cancer, normal pregnancy, complications of pregnancy (e. G., Ischemic pregnancy), myocardial infarction, systemic autoimmune disease, local autoimmune disease, inflammatory bowel disease, Allograft, < RTI ID = 0.0 > stroke, < / RTI > and localized tissue injury without rejection.

In some embodiments, the tissue or cell type is a primary tissue from a healthy subject.

In some embodiments, the tissue or cell type is an immortalized cell line.

In some embodiments, the tissue or cell type is a biopsy from a tumor.

In some embodiments, the reference map is based on sequence data obtained from samples obtained from at least one reference subject. In some embodiments, the sequence data defines the location of a cfDNA fragment end-point in the reference genome, for example, when a reference map is created by sequencing cfDNA from a subject (s) with a known disease. In another embodiment, the sequence data on which the reference map is based is based on a DNase I hypersensitive site data set, an RNA expression data set, a chromosome conformation map, or a chromosome accessibility map, or a digestion by a micrococycle nuclease And a nucleosome mapping map generated by the nucleosome mapping map. In some embodiments, the reference object is healthy. In some embodiments, the reference subject has a disease or disorder. In some embodiments, the disease or disorder is selected from the group consisting of cancer, normal pregnancy, complications of pregnancy (e. G., Ischemic pregnancy), myocardial infarction, systemic autoimmune disease, inflammatory bowel disease, local autoimmune disease, Allograft, < RTI ID = 0.0 > stroke, < / RTI > and localized tissue injury without rejection.

In some embodiments, the reference map comprises a cfDNA fragment end point probability or amount correlated with said probability, for at least a portion of the reference genome associated with a tissue or cell type. In some embodiments, the reference map comprises a mathematical transformation of the cfDNA fragment endpoint probability or amount correlated with such probability.

In some embodiments, the reference map includes a score for at least a portion of the coordinates of the reference genome associated with the tissue or cell type. In some embodiments, the reference map includes a mathematical transformation of the score, such as a Fourier transform of the score. In some embodiments, the score is based on annotations of reference genomic coordinates for tissue or cell type. In some embodiments, the score is based on the location of the nucleosome and / or the chromatome. In some embodiments, the score is based on a transcription initiation site and / or a transcription termination site. In some embodiments, the score is based on the predicted binding site of at least one transcription factor. In some embodiments, the score is based on predicted nuclease hypersensitive sites.

In some embodiments, the score is associated with at least one orthogonal biological feature. In some embodiments, orthogonal biological features are associated with highly expressed genes. In some embodiments, orthogonal biological features are associated with a low expression gene.

In some embodiments, at least some of the plurality of scores each have a score above the threshold value. In such an embodiment, the score below the threshold (min) value is excluded from comparing the score to the reference map. In some embodiments, the threshold value is determined prior to determining the tissue (s) and / or cell type (s) that produce the cfDNA. In another embodiment, the threshold value is determined after determining the tissue (s) and / or cell type (s) that produce cfDNA.

In some embodiments, determining the tissue and / or cell type that produces cfDNA as a function of a plurality of genomic locations of at least a portion of the cfDNA fragment endpoints comprises determining the type of cfDNA endpoint of the sample having one or more characteristics of the one or more reference maps And at least some of the genomic location distributions.

In some embodiments, the mathematical transform includes a Fourier transform.

In some embodiments, the reference map comprises enzyme-mediated fragmentation data corresponding to at least one tissue associated with the disease or disorder.

In some embodiments, the reference genome is human.

In one aspect of the invention, the methods described herein are used for the detection, monitoring and evaluation of tissue (s) and / or origin cell type (s) of malignant tumors from analysis of cfDNA in body fluids. It is now well established that in malignant tumor patients, some portion of cfDNA in body fluids such as circulating plasma may be derived from the tumor. The methods described herein can potentially be used to detect and quantify tumor-derived portions. In addition, since the nucleosomal occupancy map is cell type specific, the methods described herein can potentially be used to determine the tissue (s) and / or origin cell type (s) of a malignant tumor. In addition, as noted above, it has been observed that the concentration of circulating plasma cfDNA in the cancer is greatly increased and there may be an imbalance in the contribution from the tumor itself. This suggests that other tissues (eg, epilepsy, immune system) are likely to contribute to circulating plasma cfDNA during cancer. to the extent that the contribution from the other tissue to the cfDNA is consistent among the patients for the other types of cancer presented, it is possible to detect cancer using the method described above, based on a signal from the other tissue, Monitoring and / or evaluation of the tissue (s) and / or origin cell type (s).

In another aspect of the invention, the methods described herein are used for the detection, monitoring and evaluation of the originating tissue (s) and / or cell type (s) of tissue damage from the analysis of cfDNA in body fluids. Many pathological processes are expected to produce part of cfDNA in body fluids such as circulating plasma derived from damaged tissue. The methods described herein can potentially be used to detect and quantify cfDNA derived from tissue damage, including identification of related tissue and / or origin cell types. This may enable the diagnosis and / or monitoring of many other pathological processes involving myocardial infarction (acute injury of cardiac tissue), autoimmune disease (chronic injury of various tissues) and acute or chronic tissue injury.

In another aspect of the invention, the methods described herein are used to evaluate the fetal fraction of cfDNA during pregnancy and / or to improve the detection of chromosomal or other genetic abnormalities. The relatively superficial sequence determination of maternal plasma DNA fragments coupled with the nucleosome maps described above can enable a cost-effective and rapid assessment of fetal fraction in both male and female fetuses. In addition, since unequal probability can be assigned to individual sequencing determinations for the likelihood that it originated from the maternal or fetal genome, the method can detect chromosomal anomalies (e. G., Trisomy) through analysis of cfDNA in maternal body fluids The performance of the test can be improved.

In another aspect of the invention, the methods described herein are used to quantify the contribution of transplantation (autologous or allograft) to cfDNA. Current methods for early and noninvasive detection of acute allograft rejection involve sequencing the plasma DNA and identifying increased concentrations of fragments derived from the donor genome. This method relies on relatively in-depth sequencing of the fragment pool to detect, for example, 5-10% donor fractions. Methods based on nucleosome maps of donated organs can enable more sensitive evaluation using similar determinations or equivalent amounts of sequence determinations using more superficial sequence determinations. Similar to cancer, it is also possible that other cell types other than the transplant itself contribute to the cfDNA composition during transplant rejection. The graft rejection can be monitored on the basis of signals from other tissues other than the graft donor cell itself in a manner as described above, to the extent that the contribution to cfDNA from these other tissues is consistent between patients during graft rejection.

Additional embodiments of the present disclosure.

The present disclosure also provides a method of diagnosing a disease or disorder using a nucleosome reference map (s) generated from a subject having a known disease or disorder. In some embodiments, the method comprises the steps of (1) creating a reference set of nucleosomal maps, wherein each nucleosomal map comprises a defined clinical status (e.g., normal, pregnancy, cancer type A, cancer type B Etc.) from cfDNA and / or DNA derived from digestion of the chromatin of a particular tissue and / or cell type (s) from body fluids of the individual (2) comparing the nucleosome map derived from its cfDNA with a reference set of nucleosomal maps to predict the composition of the clinical state and / or tissue / origin cell type of cfDNA from the body fluid of the individual (s) do.

Step 1: Create a reference set of nucleosome maps and collect or summarize signals from the nucleosome batch.

Preferred methods for generating a nucleosome map include DNA purification, library construction (by adapter ligation and possibly by PCR amplification) and mass parallel sequencing of cf DNA from body fluids. An alternative source of nucleosome maps useful as a reference point or in terms of the present invention to identify the major components of the variant is chromatin digestion by micrococycle nuclease (MNase), DNase treatment, ATAC-Seq or distribution is DNA derived from other related methods in which information about the nucleosome arrangement in (a), (b) or (c) is collected. Descriptions of these distributions (a), (b) and (c) are provided in the paragraph number [0020] and are shown graphically in FIG.

In principle, the high degree of sequencing of these libraries can be used to quantify nucleosomal occupancy in aggregated cell types that contribute to cfDNA at specific coordinates of the genome, but this is currently very expensive. However, signals associated with the nucleosomal occupancy pattern can be summarized or aggregated across a continuous or discontinuous region of the genome. For example, in Examples 1 and 2 presented herein, the distribution, i.e. distribution (a), of the sites in the reference human genome to which the sequencing reading initiation site is mapped is determined by the Fourier transform in a contiguous window of 10 kilobase pairs (kbp) Followed by quantification of the intensity over the frequency range associated with the occupancy of the nucleosome. This effectively summarizes the degree to which the nucleosomes exhibit a structural arrangement within each 10 kbp window. In Example 3 presented herein, we have found that when TFBS is bound by TF, it is located immediately adjacent to the transcription factor binding site (TFBS) of a particular transcription factor (TF), which is often flanked directly by the nucleosome , The distribution (a) of the region in the reference human genome to which the sequence reading initiation site is mapped, that is, the distribution (a) is quantified. This effectively summarizes the nucleosome arrangement as a result of TF activity in the cell type (s) that contribute to the cfDNA. Importantly, there are many relevant ways in which the signal of occupancy of nucleosomes can be summarized meaningfully. This may be around a subset of all such sites defined by other genomic landmarks such as DNase I sensitive sites, transcription initiation sites, phase domains, other epigenetic markings, or correlated behavior (e.g., gene expression, etc.) (A), (b), and / or (c). As the sequencing cost continues to fall, it is possible to use the nucleosomal occupancy map as a reference map, i.e. without directly counting signals, including those made from cfDNA samples related to known diseases for comparison with unknown cfDNA samples . In some embodiments, the biological sample from which a reference map of nucleosomal occupancy is made is collected from an animal xenografted with a human cell or tissue. The advantage of this is that sequenced cfDNA fragments mapped to the human genome will be derived exclusively from xenografted cells or tissues, as opposed to representing a mixture of cfDNAs derived from the cell / tissue of interest with the hematopoietic lineage.

Step (2): The cfDNA-derived nucleosome map of one or more new entities / samples is compared directly or with mathematical transformation of each map to the reference set of nucleosome maps to determine the pathology (s), clinical condition (s) And / or predicting tissue / origin cell type composition.

Once a reference set of nucleosomal maps is created, there are various statistical signal processing methods for comparing additional nucleosomal map (s) with reference sets. In Examples 1 & 2, we first summarize the long range of nucleosome sequences within the 10 kbp window along the genomes of the various sample sets and then cluster the samples (Example 1) to estimate the mixture ratio (Example 2) Principal component analysis (PCA) is performed. We know the clinical status of all cfDNA samples and tissue / origin cell types of all cell line samples used in these examples, but one of the samples may in principle be "unknown " It is used to predict the presence / absence of clinical condition or its tissue / origin cell type based on its behavior in PCA analysis compared to all other nucleosomal maps.

An unknown sample does not necessarily have to be matched exactly to 1+ members of the reference set in a 1: 1 manner. Rather, it can quantify its similarity to each (Example 1), or its nucleosome map can be modeled as a non-uniform mixture of 2+ samples from a reference set (Example 2).

The tissue / origin cell type composition of cfDNA in each sample need not be predicted or ultimately known for the success of the method of the present invention. Rather, the methods described herein rely on the consistency of the tissue / origin cell type composition of cfDNA in terms of a particular disease or clinical condition. However, by analyzing DNA derived from chromatin digestion and adding it to a nucleosome map to directly examine the nucleosome maps of many tissues and / or cell types, the tissue (s) and / or tissue (s) contributing to the unknown cfDNA- Or cell type (s).

In any of the embodiments described herein, the method may further comprise the step of writing a report that includes a statement identifying the disease or disorder. In some embodiments, the report can further include a list of determined tissues and / or cell types that produce isolated cfDNA. In some embodiments, the report further comprises a list of diseases and / or disorders that are unlikely to be associated with the subject. The report may optionally include a sample and / or a tissue, a tissue that is likely not to produce any cfDNA isolated from the biological sample, the type of biological sample, the date the biological sample was obtained from the subject, the date the cfDNA isolation step was performed and / (S) and / or any other information about the cell type (s).

In some embodiments, the reports include, but are not limited to, proposals for additional diagnostic tests, proposals for initiation of treatment regimens, suggestions for alteration of existing therapies and / or proposals to delay or discontinue existing therapies, Additional recommended treatment protocols are included.

Example

Example  One. Acellular cell  DNA Nucleosomes  Principal component analysis of maps

In order to evaluate the presence of the signal associated with the nucleosome arrangement, the distribution of the reading initiation site in the sequencing data derived from the cfDNA extraction and MNase digestion experiments was examined. To this end, the collected cfDNA samples (human plasma including contributions from unknown healthy individuals; bulk.cfDNA), cfDNA samples from one healthy male control (MC2.cfDNA), cfDNA samples from intracranial tumor patients Four cfDNA samples (tumor.2349, tumor.2350, tumor.2351, tumor.2353), six MNase digestion experiments from five different human cell lines (Hap1.MNase, HeLa.MNase, HEK.MNase, NA12878.MNase , 7 HeLaS3, MCF.7) and 7 cfDNA samples (gm1matplas, gm2matplas, im1matplas, fgs002, fgs003, fgs004, fgs005) from different pregnant female individuals were analyzed and the DNA extracted from female lymphocyte lineage (NA12878) And against the common shotgun sequencing data set. A subset of the pooled cfDNA samples (26%, bulk.cfDNA_part) and one healthy male control (18%, MC2.cfDNA_part) were also included as separate samples to investigate the effect of sequence depth.

The read start coordinates are extracted and the periodicity is generated using Fast Fourier Transform (FFT) as described in the method section. This analysis determines how much nonuniformity can be accounted for by a particular frequency / periodic signal in the distribution of the read-start site. We focused on the range of 120-250 bp, including the length range of the DNA surrounding one nucleosome (147 bp) and an additional sequence of nucleosomal linker sequences (10-80 bp). Figure 3 shows the mean intensity for each frequency across all blocks of human chromosome 1 and human chromosome 22. In addition to the cfDNA sample, the MNase digestion experiment shows a clear peak under the 200 bp cycle. These peaks are not observed in human shotgun data. This analysis is consistent with the main effect of nucleosome placement on the distribution of fragment boundaries of cfDNA.

Changes in the exact peak frequency between samples were also observed. This may be the result of a distribution with possibly different linker sequence lengths in each cell type. The origin of the peak from the pattern of the nucleosome binding DNA plus linker sequence is supported by the observation that the sides around the peak are not symmetrical and the intensity to frequency higher than the peak is lower than the frequency lower than the peak. This suggests that a plot similar to that shown in Figure 3 can be used to perform quality control of cfDNA and MNase sequence determination data. Random fragmentation or contamination of cfDNA and MNase using regular (shotgun) DNA will result in dilution or extreme case complete elimination of this characteristic intensity pattern in the cycle.

Below, the data were analyzed based on the intensity measured at a cycle of 196 bp as well as all the intensity determined for the frequency range of 181 bp to 202 bp. Because a wider range of linker lengths is captured, a wider frequency range has been chosen to provide higher resolution. These intensities were mainly chosen for purely computational reasons; In a related embodiment, different frequency ranges may be used. Figures 4 and 5 explore the visualization of the periodic intensity at 196 bp over a continuous non-overlapping 10 kbp block that spans the entire length of the human autosomal chromosome (see Methods for details). Figure 4 shows the principal component analysis (PCA) of the data and the projection over the first three components. Principal component 1 (PC1) (28.1% dispersion) captures the differences in intensity magnitudes shown in FIG. 3, thus separating MNase and cfDNA samples from genomic shotgun data. In contrast, PC2 (9.7% dispersion) captures the difference between MNase and cfDNA samples. PC3 (6.4% dispersion) captures the difference between individual samples. Figure 5 shows a hierarchical clustering tangent of the data based on Euclidean distance of the intensity vector. We note that although the data were generated in accordance with different experimental protocols in different laboratories, two HeLa S3 experiments were tightly clustered in PCA and Tendogram. "Normal" cfDNA samples, tumor cfDNA samples, and cell line MNase sample clusters were also clustered. In particular, three tumor samples derived from the same tumor type (polymorphic glioblastoma) appear to be clustered separately from tumor.2351 samples derived from other tumor types (see Table 1). GM1 and IM1 samples are clustered separately from other cfDNA samples from pregnant women. This is consistent with the higher intensity observed for frequencies below the peak in these samples (i.e., the more pronounced left shoulder in FIG. 3). This may indicate subtle differences in cfDNA production between two sample sets or unmanaged biological differences (e. G., Gestational period).

Figures 6 and 7 show the analysis results based on the frequency range of 181 to 202 bp, which is equivalent. When comparing these plots, the results are largely stable over a wider frequency range, but additional frequencies can improve sensitivity in more precise scale analysis. To further explore the cell type origin specific pattern, the cfDNA and MNase data sets were analyzed separately using PCA of intensity for this frequency range. In the following analysis set, five cfDNA samples from pregnant women with a more pronounced left shoulder in Fig. 3 were excluded. Figure 8 shows the first seven major components of the cfDNA data, and Figure 9 shows all six major components for the six MNase data sets. Clustering of related samples is present, but there are also significant variations (biological and technical variations) to separate each sample from the remaining samples. For example, the effect of sequence depth was observed, as can be seen from the separation of MC2.cfDNA and MC2.cfDNA_part as well as bulk.cfDNA and bulk.cfDNA_part. Read sampling can be used to correct this technology disturbance.

Some of the key observations of this embodiment include:

1) The reading start coordinates in the cfDNA sequencing data capture the strong signal of the nucleosome arrangement.

2) The difference in signal of the nucleosome batches aggregated over a subset of genomes such as the consecutive 10 kbp window correlates with the origin of the sample.

Example  2 - Nucleosomes  Estimate mixture ratio of maps

In Example 1, basic clustering of samples created or downloaded from a public database was studied. Analysis indicates that the reading start coordinates of this data set capture a robust signal of the nucleosome arrangement (over a range of sequence depths from 20 million to over one billion sequences) and the sample origin correlates with this signal . For the purposes of this method, it would also be useful to identify mixtures of known cell types and to quantify the contribution of each cell type from this signal to some extent. For this purpose, this example investigated a synthetic mixture of two samples (i.e., based on sequence readings). The present inventors have determined that the two MNase data sets (MCF.7 and NA12878.MNase) and the two cfDNA data sets (tumor.2349 and bulk.cfDNA) are 5:95, 10:90, 15:85, 20:80, Sequence determinations were mixed at a ratio of 30:70, 40:60, 50:50, 60:40, 30:70, 80:20, 90:10 and 95: 5. The synthetic MNase mixture data set was extracted from two sets of 196.9 million ordered readings (each from one of the origin samples), and the synthetic cfDNA mixture data set contained two sets of 181 million (From one of the origin samples, respectively).

Figure 10 shows the average intensity for chromosome 11, which is equivalent to Figure 3, except for these synthetic mixtures. From Fig. 10, it can be seen how different sample contributions cause movement of the global frequency intensity pattern. This signal can be used to estimate the synthesis mixture ratio. Figure 11 shows the first two principal components for the MNase dataset mixture, and Figure 12 shows the first two principal components of the cfDNA dataset mixture. In both cases, the first PC directly captures the composition of the mixed data set. Thus, one can directly think about how a mixture ratio for two or possibly more cell types can be estimated from the conversion of the frequency intensity data given an appropriate reference set, for example using a regression model. Figure 13 shows a tandem of two sets of data confirming the overall similarity of a mixture sample derived from similar sample ratios and separation of cfDNA and MNase samples.

One of the key observations in this embodiment is that the mixture ratio of various sample types (cfDNA or cell / tissue type) to unknown samples can be estimated by modeling the nucleosomal occupancy pattern.

Example  3: cfDNA  Using the sequencing data, Nucleosomes  Measurement of Occupancy

While the previous example demonstrates that the signal of the nucleosome arrangement can be obtained by dividing the genome into a continuous and non-overlapping 10 kbp window, the orthogonal method can also be used to create a truncation accessibility map, The tendency of artifacts may be smaller based on size and boundaries. One such method that has been investigated in this embodiment in some detail is the estimation of the nucleosome placement through the observed periodicity of readout initiation around the transcription factor (TF) binding site.

It is well established that the local nucleosome arrangement is affected by the nearby TF occupancy. The effect on the local remodeling of the chromatin and the effect on the stable arrangement of the adjacent nucleosomes is not uniform across the TF set and occupancy of the presented TF is preferentially in the nucleosome arrangement located 5 ' or 3 ' Can have local effects, and can lead to larger or smaller genomic distances in certain cell types. Also for the purposes of the present disclosure, importantly, the set of TF binding sites occupied in vivo in a particular cell will vary between tissue and cell types, and thus the TF binding site occupancy map for the tissue or cell type of interest And repeating the process for one or more TFs, confirm the concentration or depletion of one or more of the cell types or tissue specific TF binding site occupancy profiles to determine the composition of the cell type and tissue mixture contributing to the population of cfDNA Can be identified.

To prove this idea, readout initiation near the TF binding site was used to visually confirm the cleaved deflection reflecting the preferential local nucleosome placement. The ChIP-seq transcription factor (TF) peak was obtained from the Encyclopedia of DNA Elements ("ENCODE") project (National Human Genome Research Institute, US National Institutes of Health, Bethesda, Md.). Genomic interval of these peaks are broad due to (average of 200 to 400 bp), the active binding site within this interval are conservative p- value cutoff (see how 1x10 ^-5, more details) and the genome for each of the binding motif Information was scanned and identified. The intersection of the two independently derived sets of predicted TF binding sites resulted in downstream analysis.

The number of readouts at each location within each candidate TF binding site of 500 bp was calculated in samples with at least 100 million sequences. Within each sample, all readouts were added at each position to yield a total of 1,014 to 1,019 positions per sample per TF, depending on the length of the TF recognition sequence.

Figure 14 shows the distribution of read initiation around 24,666 CTCF binding sites in the human genome in a variety of different samples, centered around the binding site itself. CTCF is an insulator binding protein and plays an important role in transcriptional repression. Previous studies suggest that the CTCF binding site is fixed in a local nucleosome arrangement that is positioned such that at least 20 nucleosomes are symmetrically and regularly spaced around the binding sites presented at approximate spacing of 185 bp. One notable feature common to nearly all of the samples in Figure 14 is the clear periodicity of the nucleosome arrangement located both upstream and downstream of the binding site, which indicates that the local and symmetrical effects of CTCF binding in vivo have been demonstrated in various cfDNA and & MNase-digested samples. Interestingly, the periods of the upstream and downstream peaks are not constant across a set of samples; The MNase digested samples show slightly wider spacing of the peaks compared to the binding sites, suggesting the usefulness of their duration as well as the intensity of the peaks.

Figure 15 shows the read-initiated distribution around 5,644 c-Jun binding sites. In this figure, the familiar period can be visually identified again for several samples, but the effect is not constant. Interestingly, three of the MNase digested samples (Hap1.MNase, HEK.MNase and NA12878.MNase) have a much flattened distribution indicating that the c-Jun binding site is not predominantly occupied in these cells or that the local chromatin remodeling Lt; RTI ID = 0.0 > c-Jun < / RTI > Regardless of the underlying mechanism, the observation that deflections in the local neighborhood of read initiation are different between TF and TF and between sample types suggests that reading of the nucleosomal occupancy to correlate or deconvolute the original tissue composition in the cfDNA sample Strengthen the potential role for initiation-based estimation.

Figure 16 shows the distribution of readout initiation around 4,417 NF-YB binding sites. The starting site distribution near this TF binding site shows a departure from symmetry where the downstream effect (to the right in each plot) is stronger than the upstream effect, as evidenced by a slight upward trajectory of the cfDNA sample see. In addition, the difference between the MNase digested sample and the cfDNA sample is noteworthy, while the former shows a flat profile which on average is difficult to identify the peak, while the latter has both a more clearly identifiable periodicity and more identifiable peaks.

Example  Methods 1 to 3

Clinical and Control Samples

During routine postpartum period postpartum care, whole blood was collected from pregnant women fgs002, fgs003, fgs004 and fgs005 and stored in a Vacutainer tube containing EDTA (BD). Whole blood from pregnant women IM1, GM1, and GM2 was obtained at 18, 13 and 10 weeks of gestation, respectively, and then kept in a BD incubator tube containing EDTA. Whole blood from glioma patients 2349, 2350, 2351, and 2353 was collected as part of the brain surgery procedure and stored in a Conductivity Tube (BD) containing EDTA for less than 3 hours. Whole blood from a male male control (Male Control 2) (MC2), which is a healthy adult male, was collected from a BD incubator tube containing EDTA. Between 4 and 10 ml of blood was available for each individual. Plasma was centrifuged at 1,000 x g for 10 minutes at 4 ° C, separated from whole blood, and the supernatant was collected and centrifuged again at 2,000 x g for 15 minutes at 4 ° C. The purified plasma was stored in 1 ml aliquots at -80 ° C until use.

Bulk human plasma containing contributions from an unknown number of healthy individuals was obtained from STEMCELL Technologies (Vancouver, British Columbia, Canada) and stored in 2 ml aliquots at-80 C until use.

Plasma sample treatment

Frozen plasma aliquots were thawed in a bench-top just prior to use. Circulating cfDNA was purified from 2 ml of each plasma sample using a QiaAMP Circulating Nucleic Acids kit (Qiagen, Venlo, The Netherlands) according to the manufacturer's protocol. DNA was quantified by custom qPCR assay targeting Qubit fluorescence detector (Invitrogen, Carlsbad, Calif., USA) and human Alu sequence.

MNase  digestion

Approximately 50 million cells of each cell line (GM12878, HeLa S3, HEK, Hap1) were grown using standard methods. Growth medium was aspirated and cells were washed with PBS. Cells were trypsinized, neutralized with 2x volume of CSS medium and then pelleted in conical tubes by centrifugation at 1,300 rpm for 5 minutes at 4 [deg.] C. Cell pellets were resuspended in 12 ml ice-cold PBS supplemented with 1X protease inhibitor cocktail, counted, and then pelleted by centrifugation at 1300 rpm for 5 minutes at 4 < 0 > C. The cell pellet was resuspended at a concentration of 3 million cells per ml in RSB buffer (10 mM Tris-HCl, 10 mM NaCl, 3 mM MgCl ₂ , 0.5 mM spermidine, 0.02% NP-40, 1X protease inhibitor cocktail) ≪ / RTI > and incubated on ice for 10 minutes with careful overturning. The nuclei were pelleted by centrifugation at 4 ° C and 1,300 rpm for 5 minutes. The pelleted nuclei NSB buffer to a final concentration of 15 M per ml (25% glycerol, 5 mM of MgAc _2, 5 mM HEPES, 0.08 mM EDTA, 0.5 mM's Fermi Dean, in 1 mM DTT, 1X protease inhibitor cocktail) Lt; / RTI > The nuclei were pelleted again by centrifugation at 1300 rpm for 5 min at 4 ° C and resuspended in MN buffer (500 mM Tris-HCl, 10 mM NaCl, 3 mM MgCl ₂ , 1 mM CaCl ₂ , 1X ≪ / RTI > protease inhibitor cocktail). The nuclei were split into 200 [mu] l aliquots and digested with 4 U of micrococose nuclease (Worthington Biochemical Corp., Lakewood, NJ) for 5 min at 37 [deg.] C. The reaction was quenched on ice by adding 85 [mu] l MNSTOP buffer (500 mM NaCl, 50 mM EDTA, 0.07% NP-40, 1X protease inhibitor) and incubated at 4 [deg.] C for 90 minutes with careful inversion. DNA was purified using phenol: chloroform: isoamyl alcohol extraction. The mononucleosomal fragments were size selected by 2% agarose gel electrophoresis using standard methods and quantified using a Nanodrop spectrophotometer (Thermo Fisher Scientific Inc., Waltham, Mass. ).

Preparation of sequence library

Treated with a ThruPLEX-FD or ThruPLEX DNA-seq 48D kit (Rubicon Genomics, Ann Arbor, Mich., USA) containing a series of proprietary end-point rescue, ligation and amplification reactions A sequencing library was prepared. 3.0 to 10.0 ng of DNA was used as input for all clinical sample libraries. Two bulk plasma cfDNA libraries were constructed using 30 ng of the input into each library; Each library was separately barcoded. Two libraries from MC2 were constructed using 2 ng of input into each library; Each library was separately barcoded. The library for each MNase digestive cell line was constructed using 20 ng size selected input DNA. To avoid over-amplification, library amplification of all samples was monitored by real-time PCR.

Sequencing

All libraries were sequenced on a HiSeq 2000 instrument (Illumina, Inc., San Diego, CA) using a 101 bp reading of the paired end with an index reading of 9 bp. The sequencing of a lane was performed on the collected samples fgs002, fgs003, fgs004, and fgs005 to generate a total of about 4.5x10 ⁷ read pairs per sample. Samples IM1, GM1 and GM2 were sequenced across multiple lanes to generate 1.2x10 ⁹ , 8.4x10 ⁸ , and 7.6x10 ⁷ read-out pairs, respectively. Sequencing of the lane is performed for the sample 2349, 2350, 2351 and 2353, respectively, it gave the read pair of approximately 2.0x10 ⁸ per sample. Sequencing of a lane was performed on each of the four cell line MNase-digested libraries to generate approximately 2.0 x ^{10 8} read-out pairs per library. Sequencing of the four lanes was performed on one of the two replicate MC2 libraries and three lanes on one of the two replicated bulk plasma libraries to generate a total of 10.6x10 ⁹ and 7.8x10 ⁸ read pairs, Respectively.

cfDNA  Processing of sequence determination data

DNA insert sizes for both the cfDNA and MNase libraries tend to be short (most of the data is 80 bp to 240 bp); Thus, adapter sequences were expected at the reading end of some molecules. The adapter sequences starting at the reading end were truncated and the forward and reverse reading of the pairing end ("PE") data for the short original molecule collapsed into a single reading ("SR"); At least an 11 bp reading and a redundant PE reading collapsed into SR. SRs representing more than 5 bases that are shorter than 30 bp or whose quality score is less than 10 have been discarded. The remaining PE and SR data were aligned to the human reference genome (GRCh37, 1000G Release v2) using the Rapid Sort tool (BWA-ALN or BWA-MEM). The generated Sequence Alignment / Map (SAM) format was converted to BAM (Binary Sequence Alignment / Map format) sorted using SAMtools.

Further publicly Available  data

The publicly available PE data of Hela-S3 MNase (accession no. SRR633612, SRR633613) and MCF-7 MNase experiment (accession no. SRR999659-SRR999662) were downloaded and processed as described above.

Illumina Cambridge ELTI. (ENA registration number ERR174324-ERR174329) of the CEPH family of 146 individuals NA12878 produced by the National Institutes of Health (Illumina Cambridge Ltd., Essex, UK). This data determined the PE sequence with a 2x101 bp reading on an Illumina HiSeq platform and the library was selected for a longer insert size before sequencing. Thus, no adapter sequence was expected at the end of the readout, so this data was directly aligned using BWA-MEM.

Extraction of Final Information

The PE data provides information on the two physical ends of the DNA molecule used to prepare the sequencing library. This information was extracted using the SAMtools application programming interface (API) in the BAM file. Both outlier coordinates of the PE data with both readings aligned on the same chromosome and the readout in opposite orientation were used. For untrimmed SR data, only one read end provides information about the physical end of the original DNA molecule. When the reading was aligned on the positive strand of the reference genome, the leftmost coordinate was used. If the reading is aligned on the reverse strand, its rightmost coordinate is used instead. When the PE data is converted to single read data by adapter trimming, both end coordinates are considered. When at least five adapter bases were trimmed in the SR sequencing experiments, both end coordinates were considered.

For the human chromosome (chromosome 1 to 22) of the human reference sequence, the number of reading ends and the application range at all positions in a window (block) of 10,000 bases were extracted. If there was no aligned reading in the block, the block was considered empty for that particular sample.

Flat periodicity

The ratios of read initiation and coverage were calculated for each non-empty block of each sample. If the coverage is zero, the ratio is set to zero. Using this ratio, the periodicity of each block was calculated using fast Fourier transform (spec.pgram of FFT, R statistical programming environment) with a frequency of 1/500 base to 1/100 base. Optionally, remove the trend (eg, subtracting the average of the series and removing the linear trend) by flattening the data (3 bp Daniell smoother; moving average giving a weight of 1/2 the final value) Parameter is used. Strengths for the frequency range of 120-250 bp were stored for each block.

Average chromosome concentration

For the sample set, non-empty blocks were identified across all samples. The intensity for a particular frequency was averaged over all blocks of each sample for each autosomal.

Principal component analysis Tendogram

Non-empty blocks were collected over the sample. Principal component analysis (PCA; R statistical programming environment prcomp) was used to reduce the number of dimensions of data and to show them in two-dimensional space. The PCA identifies the dimension that captures the greatest change in data and builds an orthogonal dimension to explain the reduction in data variation.

The Euclidean distance for each pair between sample intensities was calculated and visualized as a ternogram (stats library of the R statistical programming environment).

Prediction of transcription factor binding site

Estimated transcription factor binding sites obtained from analysis of ChIP-seq data generated across multiple cell types were obtained from the ENCODE project.

An independent set of candidate transcription factor binding sites was obtained by scanning the human reference genome (GRCh37, 1000G release v2) using the fimo program from the MEME software package (version 4.10.0_1). Scans were performed using the position weight matrix obtained from the JASPAR_CORE_2014_vertebrates database using the "--verbosity 1 --thresh 1e-5" option. The transcription factor motif identifiers used were MA0139.1, MA0502.1 and MA0489.1.

The chromosome coordinates from both sets of predicted sites were crossed with bedtools v2.17.0. To preserve any asymmetry in the plot, only the predicted binding sites on the "+" strand were used. Read initiation was counted for each sample within 500 bp at either end of the predicted binding site and was summed in the sample by location across all these sites. Only at least samples with a total reading of 100 million were used in this analysis.

Example  4: From cfDNA  Determination of Origin / Healthy Organization (s)

This includes evidence of genomic organization of the cells that produce these fragments, and thus of the originating tissue (s) of the cf DNA molecule population, even when there is no genotype difference between the fragment types observed in the cfDNA of a single entity, even contributing cell types , CfDNA was sequenced in-depth to better understand the process of generating it. The data generated was used to construct a map of the entire genome of nucleosomal occupancy based on previous studies by other investigators, but is considerably more comprehensive. By optimizing the library manufacturing protocol to recover short fragments, we have found that the in vivo occupancy of transcription factors (TF) such as CTCF is directly tracked by cfDNA. Finally, as revealed by cfDNA sequencing in healthy individuals, the regulatory elements and the nucleosome spacing of the gene body were found to be most closely related to DNase hypersensitivity and gene expression in lymphoid and myeloid cell lines.

cfDNA  Short Chromatochem  Corresponding and substantial DNA damage Include

Conventional sequencing libraries were prepared by end-repair and adapter ligation of purified cfDNA fragments from plasma from plasma or single ("IH01") pools from unknown numbers of healthy individuals ("BH01" 17: Table 1):

<Table 1>

Sequence statistics for plasma samples.

SSP, single-strand library manufacturing protocol. DSP, double-stranded library manufacturing protocol.

For each sample, the total number of sequenced fragments, the read length, the percentage of the fragments that align with the reference material with or without the mapping quality threshold, the average coverage, the overlap rate, and the length of the fragment The statistics related to the determination of the sequence including the ratio were tabulated. The fragment length was estimated from the alignment of the pair forming end reading. Since the read length is short, the application range is calculated assuming that the entire fragment has been read. The estimated number of redundant fragments is based on the fragment endpoint, which can overestimate the true redundancy rate in the presence of highly shaped cleavage. SSP, single-strand library manufacturing protocol. DSP, double-stranded library manufacturing protocol.

The libraries BH01 and IH01 were sequenced to 96- and 105-fold coverage (1.5G and 1.6G fragments), respectively. The fragment length distribution estimated from the alignment of the paired end reads has ~ 167 bp (consistent with the length of the DNA associated with the chromatome) and a predominant peak at ~ 10.4 bp in the 100-160 bp length range ). This distribution is consistent with the model in which the cfDNA fragment is preferentially protected from association with proteins, in this case by nuclease core particles and linker histone, by nuclease cleavage both before and after death of the protein , Some additional nicking or truncation occurs with respect to the helical pitch of the nucleosome-binding DNA. Additional support for this model reproduces the key features of the initial study on MNase-derived nucleosome-related fragments (eg, biosynthetics on A / T dinucleotides in dichromosomes) and nucleosome core particles are crossover Lt; / RTI &gt; fragment of the 167 bp fragment (Fig. 19).

The prediction for this cfDNA ontology model is extensive DNA damage, such as 5 'and 3' overhangs as well as single strand nicks. During the conventional library manufacturing process, nicked strands are not amplified, overhangs are smoothed by terminal repair, and short double-stranded DNA (dsDNA) molecules that can represent a substantial fraction of the total cfDNA can only be recovered poorly. To address this, a single stranded sequencing library from plasma containing cfDNA derived from an additional healthy individual ('IH02') was used to screen for extensive DNA damage and nucleases cleavage around the nucleosome, , And the like. Briefly, cfDNA was denatured and a biotin-conjugated single-stranded adapter was ligated to the resulting fragment. The ligated fragment was then applied to the ligation of the second strand synthesis, the end repair and the second adapter while securing the fragment to streptavidin beads. Finally, minimal PCR amplification was performed, enriching the adapter-retaining molecule, while also adding the sample index (Figure 20; Table 2).

<Table 2>

Synthetic oligos used in the manufacture of single stranded sequencing libraries.

For IH02, the generated library was sequenced to a 30-fold coverage (779M fragment). The fragment length distribution again showed a dominant peak at ~ 167 bp corresponding to the chromatogram, but was considerably enriched in shorter fragments as compared to conventional library preparation (Figures 21, 22, 23a-b, 24a-b). Although all libraries exhibit a periodicity of ~ 10.4 bp, the fragment size differs by 3 bp for both methods, indicating that its true endpoint is more accurately represented in a single-stranded library than a damaged or non-flush input molecule .

In-depth cfDNA  In vivo based on sequence determination Nucleosomes  A map of the entire genome of protection

To assess whether the significant local location of the nucleosome across the human genome in the tissue (s) contributing to the cfDNA can be estimated by comparing the distribution of aligned endpoints of the fragment or its mathematical transformation to one or more reference maps, ("WPS") was developed. Specifically, the cfDNA fragment endpoint was expected to be clustered adjacent to the nucleosome border, depleted in the nucleosome itself. To quantify this, WPS has been developed, which represents the number of fragments with endpoints in the same window minus the number of DNA fragments entirely in the 120 bp window centered on the proposed genomic coordinates (Fig. 25). As intended, the value of WPS correlates with the location of the nucleosomes in the strongly positioned array, as mapped by in vitro methods or other groups using ancient DNA (Fig. 26). At other sites, WPS correlates with genomic features such as the DNase I hypersensitivity (DHS) site (e. G., Coincident with the repositioning of nucleosomes present at the side of the distal regulatory element) (Figure 27).

Heuristic algorithms were applied to the genome-wide WPS of the BH01, IH01 and IH02 data sets to confirm 12.6M, 11.9M and 9.7M local maxima of the nucleosome protection, respectively (Figs. 25-31). In each sample, the distance distribution scheme between adjacent peaks was 185 bp with a low deviation (Fig. 30), which was generally consistent with a previous analysis of the nucleosome repeat length in human or mouse cells.

The genomic distance of each peak in the sample for the nearest peak in each different sample was calculated to determine if the position of the peak call was similar across the sample. High agreement was observed (Fig. 31; Fig. 32a-c). The median (absolute) distance from the BH01 peak call to the closest neighbor IH01 peak call was 23 bp overall, but less than 10 bp at the highest score peak (Figs. 33a-b).

The fragment endpoints were also simulated, either by nuclease specificity or because the bias introduced during library preparation could contribute artificially to the nucleosome protection signal, to match the depth, size distribution and terminal dinucleotide frequency of each sample . Then, for the simulation data sets that compute the WPS over the genome and match BH01, IH01 and IH02 respectively, 10.3 M, 10.2 M, and 8.0 M by the same heuristic are referred to as local maxima. The peaks from the simulated data set were associated with a lower score than the peaks of the actual data set (Figures 33a-b). In addition, the positions of the relatively reproducible peaks mentioned from the actual data set (Fig. 31, Fig. 32a-c) were not well aligned with the positions of the peaks mentioned from the simulated data set (Fig. 31; Figs. 34a-c).

To improve the accuracy and completeness of the nucleosome maps throughout the genome, the cfDNA sequencing data from BH01, IH01 and IH02 were collected and compared to the combined 231-fold coverage ('CH01'; 3.8B fragment; Table 1) Respectively. WPS was calculated and a 12.9 M peak was mentioned for the combined sample. This peak call set was associated with a higher score and reached saturation in terms of number of peaks (Figures 33a-b). Considering all peak-to-peak distances less than 500 bp (Figure 35), the CH01 peak set spans the 2.53 base pair (Gb) of the human reference genome.

Nucleosomes are known to be well located in relation to landmarks for gene regulation, such as transcription initiation sites and exon-intron boundaries. Consistent with that understanding, similar batches were also observed in this data with respect to landmarks for transcription, translation and splicing (Figures 36-40). Based on past observations of the correlation between the nucleosomal spacing and transcriptional activity and chromatin labeling, compartment A (rich in open chromatin) or compartment (based on locus Hi-C) on a lymphocyte- Peak to peak spacing median values in 100 kilobase (kb) windows assigned to B (closed chromatin rich) were investigated. The nucleosomes of compartment A were spaced more closely than the nucleosomes of compartment B (median 187 bp (A) versus 190 bp (B)), with further differences between specific subcompartments (Figure 41) . Except for the fact that the central nucleosome spacing, which is induced by the length of the chromosome, and by the strong arrangement across the array of alpha satellites (171 bp monomer length; Fig. 42; Fig. 26) I did not see a common pattern.

short cfDNA  Short CTCF  And other transcription factors directly Track

Previous studies on the DNase I cleavage pattern identified two predominant fragment classes, a longer fragment associated with cleavage between nucleosomes and a shorter fragment associated with cleavage adjacent to the transcription factor binding site (TFBS). To evaluate whether in vivo induced cfDNA fragments also originated from two sensitivity classes for nuclease cleavage, we split the sequence readout (CH01) based on the estimated fragment length and plot the WPS against long fragments 120-180 (bp; 120 bp window; essentially the same as the WPS described above for nucleosome calling) or short fragments (35-80 bp; 16 bp window) were separately recomputed (Figures 26-27). To obtain a well-defined set of enriched TFBS for the sites actively engaged in our data, the clustered FIMO prediction was crossed with an integrated set of ChIP-seq peaks of ENCODE (TfbsClusteredV3) for each TF.

Long fraction WPS supports strong organization of nucleosomes near the CTCF binding site (Figure 43). However, a strong signal is observed in short fraction WPS, which is consistent with the CTCF binding site itself (Figs. 44-45). The CTCF binding sites were stratified based on the assumption that they are bound in vivo (an additional subset that intersects all FIMO predictions versus ENCODE ChIP-seq crossing subsets versus what appears to be used across 19 cell lines). An experimentally well supported CTCF site was found to be located between adjacent -1 and +1 nucleosomes based on long fraction WPS, consistent with its repositioning (~ 190 bp to 260 bp, Figure 45-48) upon CTCF binding Substantially wider spacing. In addition, the experimentally well-supported CTCF site displays a much stronger signal for the fractionated WPS than the CTCF binding site itself (Figures 49-52).

FIMO prediction and ENCODE CHiP-seq data both performed a similar analysis for the additional TF available (Figs. 53a-h). For many of these TFs, such as ETS and MAFK (Figs. 54-55), a short fraction footprint was observed and a periodic signal was involved in long fraction WPS. This is consistent with the strong arrangement of nucleosomes surrounding the combined TFBS. Overall, these data support the view that short cfDNA fragments recovered significantly better by the single-stranded protocol (Figure 18, Figure 21) directly track in vivo occupancy of DNA-binding transcription factors including CTCF and the like.

Nucleosomes  The spacing pattern is cfDNA  Information about the origin organization to provide

To determine if in vivo nucleosome protection as measured through cfDNA sequencing can be used to estimate the cell types contributing to cfDNA in healthy individuals, the ability of the nucleosomes in the DHS region defined in 116 diverse biological samples The peak to peak spacing was investigated. The enlarged spacing may have been previously (e.g., anecdotally at the DHS site (Figure 27) or at the combined CTCF site (Figure 45)) -1 and +1 nucleosomes Respectively. Similar to the bound CTCF site, a substantially wider gap was observed for the nucleosome pair within a subset of the DHS sites, and this is reasonably related to the intervening transcription factor binding in the cell type (s) that produce cfDNA Corresponds to the site at which the nucleosome is relocated (~ 190 bp to ~ 260 bp, Figure 56). In fact, the ratio of extended nucleosome spacing (~260 bp) is significantly different depending on which cell type of DHS site is used. However, all cell types with this highest ratio are lymph or bone marrow origin (e. G., CD3_CB-DS17706 in Figure 56). This is consistent with hematopoietic cell death as a major source of cfDNA in healthy individuals.

Next, the nucleosome protection signal was re-examined near the transcription initiation site (Figure 36). When the signal was stratified based on gene expression in the lymphoid lineage cell line NB-4, strong differences in the location or intensity of nucleosome protection in relation to TSS were observed in highly expressed versus low expressed genes 57). In addition, the short fraction WPS shows a clear footprint just upstream of the TSS whose intensity is also strongly correlated with the expression level (Figure 58). This reflects the footprint of the transcriptional pre-initiation complex or some of its components in the transcriptional activation gene.

These data demonstrate that the cfDNA fragmentation pattern actually contains signals that can be used to estimate the tissue (s) or cell type (s) that produce the cfDNA.

However, the problem is that relatively few readings in the cfDNA library across genomes directly overlap with the DHS and transcription initiation sites.

The nucleosome spacing is different between cell types, and as a function of chromatin state and gene expression. In general, open chromatin and transcription are associated with shorter nucleosome repeat lengths, which is consistent with the analysis of this example of segment A versus B (Figure 41). The peak call data in this example also indicates a correlation between the expression level of the nucleosomes and the interval between the nucleosomes over the gene body and a tighter interval is associated with higher expression (Figure 59: ρ = -0.17; n = 19,677 gene). The correlation is highest for the gene itself compared to the adjacent region (upstream 10 kb ρ = -0.08; downstream 10 kb ρ = -0.01). If the assay is restricted to the gene body spanning at least 60 nucleosome calls, the more dense nucleosome spacing has a much stronger correlation with gene expression (p = -0.50; n = 12,344 genes).

One advantage of using signals such as the nucleosome spacing across the gene body or other domains is that a much greater proportion of the cfDNA fragments will provide the nucleotide. Another potential benefit is the ability to detect mixtures of signals that occur in different cell types that contribute to cfDNA. To test this, additional mathematical transformations, i.e., fast Fourier transforms (FFT), were performed on the long fragment WPS over the first 10 kb gene body and on a gene unit basis. The intensity of the FFT signal was correlated with gene expression in a specific frequency range, with a positive correlation being the highest at 177-180 bp and a negative correlation at ~ 199 bp (Figure 60). When performing the above analysis on 76 expression data sets for human cell lines and primary tissues, the strongest correlation was observed in the hematopoietic lineage (Figure 60). For example, for each of the three healthy samples (BH01, IH01, IH02), the highest order of voice correlation with average intensity in the 193-199 bp frequency range was found in lymphocyte lines, bone marrow cell lines or bone marrow tissues 61; Table 3).

<Table 3>

Correlation of gene expression data sets with WPS FFT intensity.

At 193-199 bp frequencies in the first 10 kb downstream of the transcription initiation site with FPKM expression values measured for 19,378 Ensembl gene identities in 44 human cell lines and 32 primary tissues by Human Protein Atlas Correlation FFT (Fast Fourier Transform) strength value for the average. Table 3 includes a brief description of each of the expression samples provided in Protein Atlas and the rank conversion and ranking differences for the IH01, IH02 and BH01 samples.

Example  5: From cfDNA  Determination of Non-Healthy Origin Organization

To test whether additional contributing tissues in the unhealthy state can be estimated, the sequence of the cfDNA samples from five late cancer patients was determined. The nucleosome gap pattern in these samples represents an additional contribution to cfDNA that has the strongest correlation with non-hematopoietic tissue or cell line often consistent with the anatomical origin of the patient's cancer.

Cancer patient From cfDNA Nucleosomes  The interval confirms the non-hematopoietic contribution.

To determine if a signature of a non-hematopoietic line that contributes to circulating cfDNA in an unhealthy state could be detected, 44 plasma samples from individuals clinically diagnosed with various IV cancers were treated with a single strand made from cfDNA (Table 4, median 2.2-fold coverage): &lt; RTI ID = 0.0 &gt;

<Table 4>

Clinical diagnosis and cfDNA yield for cancer panel.

§: Samples were selected for further sequencing.

^** Only 0.5 ml of plasma could be used for this sample.

†: The sample failed the QC and was not used for further analysis.

Table 4 shows the clinical and histological diagnosis of 48 patients screened for plasma-containing cfDNA for evidence of high tumor burden, with total cfDNA yields from 1.0 ml plasma from each individual and associated clinical covariates . Of these 48, 44 passed the QC and had enough material. Of these 44, 5 were selected for in-depth sequencing. The yield of cfDNA was determined by Qubit Fluorescence Meter 2.0 (Life Technologies).

These samples were produced in the same batch with the same protocol as IH02 of Example 4. Human peripheral blood plasma for 52 individuals clinically diagnosed with IVC cancer (Table 4) was obtained from Conversant Bio or PlasmaLab International (Everett, Wash., USA) and was maintained at -80 &lt; 0 &gt; C In 0.5 ml or 1 ml aliquots. Human peripheral blood plasma for four individuals clinically diagnosed with systemic lupus erythematosus was obtained from Converted Biotech and stored in 0.5 ml aliquots at -80 ° C until use. Frozen plasma aliquots were thawed in a bench-top just prior to use. Circulating acellular DNA was purified from 2 ml of each plasma sample using a QiaAMP circulating nucleic acid kit (quiagen) according to the manufacturer's protocol. DNA was quantitated with a Qubit fluorescence analyzer (Invitrogen). To confirm the yield of cfDNA in a subset of samples, the purified DNA was further quantified by custom qPCR assays targeting multiple copies of the human Alu sequence; Both estimates were found to be consistent.

Since the matched tumor genotypes were not available, each sample was scored with two metrics of the following arealities to identify a subset likely to contain a high percentage of tumor-derived cfDNA: First, the deviation from the predicted rate of readings derived from each chromosome (Fig. 62A); Second, an allelic genotype profile per chromosome for a panel of common single nucleotide polymorphisms (Figure 62b). Based on these metrics, a single-stranded library derived from 5 individuals (with small cell lung cancer, squamous cell lung cancer, colorectal adenocarcinoma, hepatocellular carcinoma and carcinoma in situ carcinoma breast cancer) was sequenced to an extent similar to that of IH02 of Example 4 (Table 5, average 30-fold coverage):

<Table 5>

Sequence statistics for additional samples included in CA01 set

SSP, single-strand library manufacturing protocol. DSP, double-stranded library manufacturing protocol. † Samples were previously published (J.O. Kitzman et al., Science Translational Medicine (2012)).

Table 5 shows the total number of sequenced fragments for each sample, the read length, the percentage of the fragments that align with reference material with or without the mapping quality threshold, the average coverage, the overlap rate, The statistics related to the determination of the sequence including the ratio of fragments are presented in the table. The fragment length was estimated from the alignment of the pair forming end reading. Since the read length is short, the application range is calculated assuming that the entire fragment has been read. The estimated number of redundant fragments is based on the fragment endpoint, which can overestimate the true redundancy rate in the presence of highly shaped cleavage.

As described above, FFT was performed on long fragment WPS values across the gene body and the average intensity in the 193-199 bp frequency range was related to the same 76 expression data sets for human cell lines and primary tissues. In contrast to the three samples from the healthy individuals of Example 4 (all the top 10 and almost all of the top twenty correlations are for the lymphatic or myeloid lineages), a large number of the highest ranking cell lines or tissues , And in some cases non-hematopoietic lines aligned with cancer types (Figure 61; Table 3). For example, in the case of IC17 in which the patient had hepatocellular carcinoma, the highest correlation was for HepG2, a hepatocellular carcinoma cell line. In the case of IC35 in which the patient had an intraepithelial carcinoma breast cancer, the highest correlation was for the metastatic breast adenocarcinoma cell line MCF7. In other cases, the cell line or primary tissue showing the greatest change in the correlation rank was aligned with the cancer type. For example, the largest change in the correlation rank (-31) was for the small cell lung cancer cell line (SCLC-21H) in the case of IC15 in which the patient had small cell lung cancer. In the case of IC20 (lung squamous cell carcinoma) and IC35 (colorectal adenocarcinoma), there were a number of non-hematopoietic cancer cell lines replacing the lymph / bone marrow cell line in terms of correlation coefficient, Respectively. It is possible that the specific molecular profiles of these cancers are not well represented among the 76 expression data sets (for example, none of them are lung squamous cell carcinomas; CACO-2 is a cell line derived from colorectal adenocarcinomas, It is known to be highly heterogeneous).

A greedy iterative method was used to assess the proportion of various cell types and / or tissues that contribute to cfDNA derived from biological samples. First, his reference map (defined here by 76 RNA expression data sets) has the highest correlation with the mean FFT intensity at the 193-199 bp frequency of the WPS long fragment value across the gene body for the presented cfDNA sample Cell type or tissue was identified. Next, a series of "two tissue" linear mixture models were fitted, including each cell type or tissue having the highest correlation, as well as each other cell type or tissue from the entire set of reference maps. In the latter set, the cell type or tissue with the highest count was maintained as a contributing factor if the count was not less than 1%, where the procedure was terminated and the final tissue or cell type was not included. This procedure was repeated with "3-Tissue", "4-Tissue", etc., until terminated, based on the newly added tissue estimated to contribute less than 1% by the mixed model. The mixed model takes the following form:

(a-log2Exp organization 1 + b * log2 organization 2 + c * log2 organization 3 + ... + (1-abc- ...) * log2Exp organization N).

For example, in the case of IC17, a cfDNA sample derived from a patient with advanced hepatocellular carcinoma, the procedure was Hep_G2 (28.6%), HMC.1 (14.3%), REH (14.0%), MCF7 (10.7%), THP.1 (7.4%), NB.4 (5.5%), U.266.84 (4.5%) and U.937 (2.4%). In the case of BH01, which is a cfDNA sample corresponding to a mixture of healthy individuals, the procedure was performed using bone marrow (30.0%), NB.4 (19.6%), HMC.1 (13.9%), U.937 (13.4%), U.266.84 (12.5%), Karpas.707 (6.5%), and REH (4.2%). Notably, in the case of IC17, a sample derived from a cancer patient, the highest predicted contribution rate corresponds to a cell line closely associated with the type of cancer (Hep_G2 and hepatocellular carcinoma) present in the patient from which this cfDNA is derived. In contrast, in the case of BH01, this method predicts the contribution of a healthy individual to the major source of plasma cfDNA, primarily tissue or cell types associated with hematopoiesis.

Example  6: Example  General method of 4-5

Sample

Bulk human peripheral blood plasma containing an unknown number of healthy donors was obtained from Stem Cell Technologies (Vancouver, BC, Canada) and stored in 2 ml aliquots at-80 C until use. Individual human peripheral blood plasma of an anonymous healthy donor was obtained from Converged Bio (Huntsville, Ala., USA) and stored in 0.5 ml aliquots at -80 ° C until use.

Whole blood from pregnant women IP01 and IP02 was obtained at 18 and 13 weeks of gestation, respectively, and treated as previously described ⁴¹ .

Human peripheral blood plasma for 52 individuals clinically diagnosed with Conservative Biotechnology or PlasmaLab International (Everett, Wash., USA) (Table 4) was obtained, and 0.5 ml at -80 ° C until use And stored in 1 ml aliquots. Human peripheral blood plasma for four individuals clinically diagnosed with systemic lupus erythematosus was obtained from Converted Biotech and stored in 0.5 ml aliquots at -80 ° C until use.

Plasma sample treatment

Frozen plasma aliquots were thawed in a bench-top just prior to use. Circulating acellular DNA was purified from 2 ml of each plasma sample using a QiaAMP circulating nucleic acid kit (quiagen) according to the manufacturer's protocol. DNA was quantitated with a Qubit fluorescence analyzer (Invitrogen). To confirm the yield of cfDNA in a subset of samples, the purified DNA was further quantified by custom qPCR assays targeting multi-copy human Alu sequences; Both assessments were found to be in agreement.

Preparation of double-stranded sequencing library

The bar-coded sequencing libraries were prepared with ThruPLEX-FD or ThruPLEX DNA-seq 48D kit (Rubicon Genomics) containing a proprietary series of end-point recovery, ligation and amplification reactions. 0.5 ng to 30.0 ng of cfDNA was used as the input of all clinical sample libraries. Library amplification for all samples was monitored by real-time PCR to avoid over-amplification and was typically terminated after 4-6 cycles.

Preparation of Single Strand Sequence Library

Adapter 2 was prepared by combining 4.5 μl TE (pH 8), 0.5 μl 1M NaCl, 10 μl 500 μM Oligo Adapter 2.1, and 10 μl 500 μM Oligo Adapter 2.2, incubated at 95 ° C for 10 seconds and at a rate of 0.1 ° C / s Lt; RTI ID = 0.0 &gt; 14 C. &lt; / RTI &gt; Refined cfDNA fragment 2x CircLigase II buffer (Epicenter (Epicentre)), 5 mM MnCl _2, and 1U alkaline phosphatase FastAP (Thermo Fisher) a combination of 0.5-10 ng fragment and a 20 ㎕ reaction volume, for 30 minutes And dephosphorylated by incubation at 37 ° C. The fragments were then denatured by heating at 95 DEG C for 3 minutes and immediately transferred to an ice bath. The reaction was supplemented with biotin-conjugated adapter Oligo CL78 (5 pmol), 20% PEG-6000 (w / v) and 200 U CircLigase II (EpiCenter) for a total volume of 40 μl, incubated with rotation at 60 ° C , Heated at 95 DEG C for 3 minutes, and then placed in an ice bath. For each sample, 20 [mu] l MyOne C1 beads (Life Technologies) were mixed with bead binding buffer (BBB) (10 mM Tris-HCl pH 8, 1 M NaCl, 1 mM EDTA pH 8, 0.05% Tween- And 0.5% SDS) and resuspended in 250 [mu] l BBB. The adapter-ligated fragment was allowed to bind to the bead by spinning at room temperature for 60 minutes. The beads were collected on a magnetic rack and the supernatant was discarded. The beads were washed once with 500 μL Wash Buffer A (WBA) (10 mM Tris-HCl [pH 8], 1 mM EDTA [pH 8], 0.05% Tween-20, 100 mM NaCl, 0.5% SDS) And washed once with buffer B (WBB) (10 mM Tris-HCl [pH 8], 1 mM EDTA [pH 8], 0.05% Tween-20, 100 mM NaCl). The beads were combined with 1X isothermal amplification buffer (NEB), 2.5 μM oligo CL9, 250 μM (each) dNTP and 24 U Bst 2.0 DNA polymerase (NEB) in a reaction volume of 50 μl and incubated at 1 ° C / Min to 15 &lt; 0 &gt; C to 37 &lt; 0 &gt; C, gently incubated and maintained at 37 [deg.] C for 10 minutes. After collection on a magnetic rack, the beads were washed once with 200 [mu] l WBA, resuspended in 200 [mu] l of stringent wash buffer (SWB) (0.1X SSC, 0.1% SDS) and incubated at 45 [deg.] C for 3 min. The beads were pooled and washed once with 200 [mu] l WBB. The beads were then incubated with 1X CutSmart buffer (NEB), 0.025% Tween-20, 100 [mu] M (each) dNTP and 5U T4 DNA polymerase (NEB) combined with gentle shaking for 30 minutes at room temperature. The beads were washed once with each WBA, SWB and WBB as described above. The beads were then mixed with 1X CutSmart buffer (NEB), 5% PEG-6000, 0.025% Tween-20, 2 μM double-stranded adapter 2 and 10 U T4 DNA ligase (NEB) Lt; / RTI &gt; The beads were washed once with each WBA, SWB and WBB as described above and resuspended in 25 μl TET buffer (10 mM Tris-HCl [pH 8], 1 mM EDTA [pH 8], 0.05% Tween-20) . The second strand was eluted from the bead by heating to 95 ° C, the beads were collected on a magnetic rack, and the supernatant was transferred to a new tube. Library amplification for all samples was monitored by real-time PCR to prevent over-amplification and required an average of 4-6 cycles per library.

Sequencing

All libraries were sequenced on a HiSeq 2000 or NextSeq 500 instrument (Illuminator).

First-order sequencing data processing

Bar-coded pairing end (PE) illuminin sequencing data was split to allow for one substitution in the bar code sequence. Readings with the same or shorter read length are referred to as consensus, and the adapter has been trimmed. The remaining consensus single-ended reads (SR) and individual PE reads were generated using the human reference genome sequence (ftp://ftp.1000genomes.ebi.ac.uk/vol1/ftp/technical) using the ALN algorithm run on BWA v0.7.10 see GRCh37, 1000 Genomes two-step technology, downloaded from / reference / phase2_reference_assembly_sequence /). PE readings were further processed with BWA SAMPE to resolve the ambiguity of the read pair or to remedy the missing alignment by a more sensitive alignment step around the position of one disposed read end. Aligned SR and PE data was converted directly into the BAM format, which was classified using the SAMtools API. The BAM file of the sample was merged over the lane and sequencing run.

Quality control was performed using FastQC (v0.11.2), and library complexity assessment (Picard tools v1.113) was obtained and the ratio of adapter dimer, analysis of estimated library insert size, nucleotides at the end of external readout and di In addition to determining the nucleotide frequency, the mapping quality distribution of each library was examined.

Simulated  Read data set

Aligned sequencing data was simulated for all major chromosomes of the human reference material (GRC37h) (SR if shorter than 45 bp, PE if not shorter than 45 bp). For this purpose, the frequency of the dinucleotides was determined from the actual data for both read ends and both strand orientations. The frequency of the dinucleotides was also recorded for the reference genome in both strands. In addition, the insert size distribution of the actual data was extracted for the 1-500 bp range. The reading was repeatedly simulated through the sequence of the major reference chromosomes. (1) randomly selecting the strands, and (2) selecting the reference sequence (s) at random to determine whether the initiation dinucleotide is taken into account (3) the insert size is sampled from the provided insert size distribution, and (4) the random number is used to determine whether the generated alignment is reported The frequency ratio of the terminal dinucleotide is used. The simulated coverage was consistent with the coverage of the original data after removal of the PCR replica.

Coverage , Read Initiation, and Windows Protection Score

The data in this disclosure provides information on the two physical ends of the DNA molecule used to prepare the sequencing library. We use the SAMtools application programming interface (API) to extract the information from the BAM file. When the reading is started, the present inventors use the externally aligned coordinates of the PE data in which the two readings are aligned on the same chromosome and the reading has the opposite direction. When the PE data is converted to single read data through adapter trimming, we regard both end coordinates of the SR alignment as read out portions. For coverage purposes, the present inventors consider all positions between two (deduced) molecular ends, including these terminal positions. We define the window protection score (WPS) of the window size k as the value obtained by subtracting the number of molecules starting from any base included in the window from the number of molecules across the window. The present inventors assign the determined WPS to the center of the window. For molecules in the 35-80 bp range (short fractions), we use a window size of 16 and for molecules of 120-180 bp (long fractions) we use a window size of 120.

Nucleosomes  peak Calling

The local maxima of the nucleosome protection are determined by the present inventors using a Savitzky-Golay filter (window size 21, quadratic polynomial) to localize to the running median value of zero (1 kb window) Long WPS. Subsequently, the WPS track is subdivided into regions of zero (up to five consecutive positions are allowed below 0). When the generated region is 50-150 bp in length, the present inventors confirm the intermediate value of the corresponding region and search for a continuous window of the maximum sum exceeding the intermediate value. We report the start, end, and center coordinates of this window. Peak-to-peak distance, etc. are calculated from the center coordinates. The score of the call is determined as the distance between the maximum value of the window and the average of the two adjacent WPS minimums neighboring the area. If the identified region is 150-450 bp in length, we apply a consecutive approach that exceeds the same median, but reports only windows that are 50-150 bp in size. For the calculation of the scores of several windows derived from the 150-450 bp region, we assume a neighboring minimum in the region to be zero. The present inventors discard regions longer than 50 bp and longer than 450 bp.

The dinucleotide composition of the 167 bp fragment

Fragments with an estimated length of exactly 167 bp, corresponding to the dominant peak of the fragment size distribution, were filtered in the sample to remove duplicates. The frequency of the dinucleotides is determined in a strand-aware manner using a sliding 2 bp window and a reference allele at each position, starting at 50 bp upstream of one fragment end and ending 50 bp downstream of the other end point Respectively. The observed frequency of dinucleotides at each position was compared to the expected frequency of dinucleotides determined from the simulated read set reflecting the same cut bias calculated in a library-specific manner (see above for details).

Surrounding the transcription factor binding site WPS  Profiles and Genome Features

The analysis started with an initial set of clustered FIMO (motif-based) intervals that defined the set of predicted transcription factor binding sites by the computer. A subset of clustered transcription factors (AP-2-2, AP-2, CTCF_Core-2, E2F-2, EBF1, Ebox-CACCTG, Ebox, ESR1, ETS, IRF-2, IRF- For the MEF2A-2, MEF2A, MYC-MAX, PAX5-2, RUNX2, RUNX-AML, STAF-2, TCF-LEF, YY1, the set of sites was actively linked Lt; RTI ID = 0.0 &gt; region. &Lt; / RTI &gt; To this end, only the predicted binding sites overlapping the peaks defined by the ChIP-seq experiment from publicly available ENCODE data (TfbsClusteredV3 set downloaded from UCSC) were retained.

The window protection scores surrounding these sites were extracted for both the CH01 sample and the corresponding simulations. The protection score for each site / feature was calculated and aggregated at each location relative to the starting coordinates of each site. The plot of the CTCF binding site shifted so that the 0 coordinate on the x-axis was located at the center of the known 52 bp binding footprint of the CTCF. The average of the first and last 500 bp (representing predominantly flat and average offset) of the 5 kb extracted WPS signal is then subtracted from the original signal. For long fragment signals only, a sliding window average was calculated using a 200 bp window and subtracted from the original signal. Finally, the modified WPS profile for simulation was subtracted from the modified WPS profile for CH01 to modify the signal resulting from fragment length and ligation bias. This final profile was plotted and referred to as "adjusted WPS ".

Genome characteristics such as transcription initiation site, transcription termination site, initiation codon, splice donor and splice acceptor site were obtained from Ensembl Build version 75. The adjusted WPS surrounding these features was calculated and plotted against the transcription factor binding site as described above.

CTCF  Around the junction Nucleosomes  Interval and corresponding WPS  analysis

The CTCF sites used in this analysis first included clustered FIMO predictions (predicted by the computer via motifs) of CTCF binding sites. We then created two additional subsets of the set as follows: 1) the intersection with the CTCF ChIP-seq peak set available via ENCODE TfbsClusteredV3 (see above), and 2) Crossing with a set of CTCF sites experimentally observed to have activity.

The positions of the ten nucleosomes on either side of the binding site were extracted for each site. We calculated the distance between all adjacent nucleosomes to obtain a nucleosomal distance distribution for each set of sites. The distribution of the nucleosome interval from -1 to +1 varied considerably, especially at larger intervals in the 230-270 bp range. This suggests that the CTCF region that exhibits the actual activity migrates mainly at wider intervals between the -1 and +1 nucleosomes, and therefore the difference in WPS for both the longer and shorter reading fractions may be evident. Thus, the average WPS of the short and long fragments at each location relative to the center of the CTCF region was further calculated. To investigate the effect of the nucleosome interval, the average was found to be less than 160 bp, 160-200, 200-230, 230-270, 270-420, 420-460 bp, and -1 to +1 nucleobases Were taken within the empty space of the cotton pad. These intervals capture the interval of interest, such as the dominant peak for the region that appears more active and the peak at 230-270 bp.

DNase  I irritable region ( DHS ) Analysis

349 of the BED format by Maurano et al. (Science, vol. 337 (6099), pp. 1190-95 (2012) and lastly modified "all_fdr0.05_hot" file on February 13, 2012) DHS peaks for primary tissue and cell line samples were downloaded from the University of Washington Encode database. Samples derived from fetal tissues, including 233 of these peak sets, were removed from the assay when they behaved inconsistently within the tissue type, possibly because multiple cell types were presented unevenly within each tissue sample. 116 samples representing various cell lines were stored for analysis. For the midpoint of each DHS peak in a particular set, the nearest upstream and downstream calls of the CH01 callset were identified and the genome distance between the centers of the two calls was calculated. The distribution of all of these distances was visualized for each DHS peak call set using a smoothed density estimate computed over a distance of 0-500 bp.

Gene expression analysis

In this study, FPKM expression values measured for 20,344 Ensembl gene identifiers in 44 human cell lines and 32 primary tissues were used by human protein atlas ("ma.csv" file). For analysis across tissues, genes with a non-zero expression value of less than 3 were excluded (19,378 genes pass through this filter). For the expression data set, the precision of the first digit below the decimal point was given for the FPKM value. Thus, the zero representation value (0.0) represents a representation between 0 and a value less than 0.05. Unless otherwise noted, the minimum expression value was set to 0.04 FPKM before the log ₂ - conversion of the expression value.

Planarity of planarity and trajectory

The long fraction WPS was used to calculate the periodicity of the genomic region using fast Fourier transform (spec.pgram in FFT, R statistical programming environment) with a frequency between 1/500 and 1/100 base. Parameters to flatten the data (moving average, 3 bp Daniel flattening agent, weighting the final value) (for example, subtracting the mean of the series and eliminating the linear tendency) may optionally be used do.

If indicated, recursive time-series filters run in the R statistical programming environment were used to remove high frequency fluctuations from the trajectory. Twenty-four filter frequencies (1 / seq (5,100,4)) were used and the first 24 values of the trajectory were used as initial values. The adjustment to the 24-value movement of the generated trajectory was made by repeating the last 24 values of the trajectory.

FFT  Correlation between Strength and Expression Value

The intensity values determined from flat periodic curves (FFT) were analyzed in relation to gene expression for the 120-280 bp range. S-shaped Pearson correlation between the gene expression value and the FFT intensity around the major nucleosome distance peak was observed. A clear negative correlation was observed in the 193-199 bp range. As a result, the average intensity in the above frequency range was correlated with the log ₂ -transformed expression value.

Further embodiments

Example 7. Isolation of cell-free DNA (cfDNA) from a biological sample from a subject, wherein the isolated cfDNA comprises a plurality of cfDNA fragments;

Determining a sequence associated with at least a portion of a plurality of cfDNA fragments;

determining the genomic location in the reference genome for at least a portion of the cfDNA fragment endpoints of the plurality of cfDNA fragments as a function of the cfDNA fragment sequence; And

determining at least a portion of the tissue and / or cell type that produces the cfDNA fragment as a function of the genomic location of at least a portion of the cfDNA fragment endpoint

Gt; and / or &lt; / RTI &gt; the cell type that produces the cfDNA in the subject.

Example 8. The method of embodiment 7 wherein the step of determining at least a portion of the tissue and / or cell type that produces the cfDNA fragment comprises comparing the genomic location of at least a portion of the cfDNA fragment endpoint with one or more reference maps / RTI &gt;

Example 9. In Example 7 or Example 8, the step of determining at least a portion of the tissue and / or cell type that produces the cfDNA fragment comprises performing a mathematical transformation on the distribution of the genomic location of at least a portion of the cfDNA fragment endpoint &Lt; / RTI &gt;

[0051] 10. The method of embodiment 9 wherein the mathematical transform comprises a Fourier transform.

11. The method of any one of embodiments 7-10, further comprising the step of determining a score for each of at least some of the coordinates of the reference genome, wherein the score comprises at least a plurality of cfDNA fragment endpoints, Wherein the step of determining at least a portion of the tissue and / or cell type that produces the observed cfDNA fragment and is determined as a function of genome location comprises comparing the score with one or more reference maps.

12. The method of embodiment 11 wherein the score for the coordinates indicates or is related to the probability that the coordinates are in the position of the cfDNA fragment end point.

13. The method of any one of embodiments 8-12, wherein the reference map comprises a DNase I hypersensitive site data set generated from at least one cell type or tissue.

Embodiment 14. The method as in any one of embodiments 8-13, wherein the reference map comprises an RNA expression data set generated from at least one cell type or tissue.

Example 15. The method of any one of embodiments 8-14 wherein the reference map is generated from cfDNA from an animal xenograft or cell xenografted.

[0063] 16. The method as in any one of embodiments 8-15, wherein the reference map comprises a chromosome stereochemistry map generated from at least one cell type or tissue.

[0061] 17. The method as in any one of embodiments 8-16, wherein the reference map comprises a chromatin accessibility map generated from at least one cell type or tissue.

18. The method as in any one of embodiments 8-17, wherein the reference map comprises sequence data obtained from a sample obtained from at least one reference object.

Embodiment 19. The method of any one of embodiments 8-18, wherein the reference map corresponds to at least one cell type or tissue associated with the disease or disorder.

Embodiment 20. The method of any one of embodiments 8-19, wherein the reference map comprises the location or spacing of nucleosomes and / or chromatomes within a tissue or cell type.

21. The method of any one of embodiments 8 to 20, wherein the reference map indicates that the chromatin obtained from at least one of the cell types or tissues is an exogenous nuclease (e.g., micrococus nuclease) . &Lt; / RTI &gt;

Embodiment 22. The method of any one of embodiments 8-21, wherein the reference map comprises chromatin accessibility data determined by a potential-based method (e.g., ATAC-seq) from at least one cell type or tissue &Lt; / RTI &gt;

[0064] 23. The method of any one of embodiments 8-22, wherein the reference map comprises data associated with DNA binding and / or location of a DNA occupancy protein for tissue or cell type.

24. The method of embodiment 23 wherein the DNA binding and / or DNA occupancy protein is a transcription factor.

Example 25. The method of embodiment 23 or 24 wherein the position is determined by chromatin immunoprecipitation of the cross-linked DNA-protein complex.

Example 26. The method of embodiment 23 or 24 wherein the location is determined by treating the DNA associated with the tissue or cell type with a nuclease (e.g., DNase-I).

Example 27. The method of any one of embodiments 8-26, wherein the reference map is a biological characteristic associated with the location or spacing of the nucleosome, chromatogram, or other DNA binding or DNA occupancy proteins in a tissue or cell type &Lt; / RTI &gt;

Example 28. The method of embodiment 27 wherein the biological characteristic is quantitative expression of one or more genes.

Example 29. The method of embodiment 27 or 28 wherein the biological characteristic is the presence or absence of one or more histone marks.

Example 30. The method of any one of embodiments 27-29 wherein the biological feature is hypersensitive to nuclease excision.

Embodiment 31. The method as in any of the embodiments 8-30, wherein the tissue or cell type used to create the reference map is a primary tissue from a subject having a disease or disorder.

Example 32. The method of embodiment 31 wherein the disease or disorder is associated with cancer, normal pregnancy, complication of pregnancy (e. G., Ischemic pregnancy), myocardial infarction, inflammatory bowel disease, systemic autoimmune disease, local autoimmune disease, Allogeneic transplantation, allografts without rejection, stroke, and local tissue injury.

[0060] Embodiment 33. The method as in any of the embodiments 8-30, wherein the tissue or cell type used to create the reference map is a primary tissue from a healthy subject.

[0215] Embodiment 34. The method as in any of the embodiments 8-30, wherein the tissue or cell type used to create the reference map is an immortalized cell line.

[0080] Embodiment 35. The method as in any of the embodiments 8-30, wherein the tissue or cell type used to create the reference map is a biopsy from the tumor.

Example 36. The method of embodiment 18 wherein the sequence data comprises the position of the cfDNA fragment end point.

Embodiment 37. The method of embodiment 36 wherein the reference object is a healthy subject.

[0215] Embodiment 38. The method of embodiment 36, wherein the reference entity has a disease or disorder.

Example 39. The method of embodiment 38 wherein the disease or disorder is associated with cancer, normal pregnancy, complication of pregnancy (e. G., Ischemic pregnancy), myocardial infarction, inflammatory bowel disease, systemic autoimmune disease, local autoimmune disease, Allogeneic transplantation, allografts without rejection, stroke, and local tissue injury.

[0080] Embodiment 40. The method as in any of the embodiments 19-39, wherein the reference map includes a reference score for at least a portion of the coordinates of the reference genome associated with the tissue or cell type.

[0099] Embodiment 41. The method of embodiment 40, wherein the reference map comprises a mathematical transformation of the score.

Example 42. The method of embodiment 40, wherein the score represents a subset of all reference genomic coordinates for tissue or cell type.

Example 43. The method of embodiment 42, wherein the subset is associated with the position or spacing of the nucleosome and / or chromatome.

Example 44. The method of embodiment 42 or 43 wherein the subset is associated with a transcription initiation site and / or a transcription termination site.

[0323] Embodiment 45. The method as in any of the embodiments 42-44, wherein the subset is associated with a binding site of at least one transcription factor.

[0215] Embodiment 46. The method as in any of the embodiments 42-45, wherein the subset is associated with a nuclease hypersensitive region.

[0099] Embodiment 47. The method as in any of the embodiments 40-46, wherein the subset is further associated with at least one orthogonal biological feature.

Example 48. The method of embodiment 47, wherein orthogonal biological features are associated with a highly expressed gene.

[0451] Embodiment 49. The method of embodiment 47, wherein orthogonal biological features are associated with a low expression gene.

[0086] 50. The method as in any of the embodiments 41-49, wherein the mathematical transform includes a Fourier transform.

[0099] Embodiment 51. The method as in any of the embodiments 11-5, wherein a subset of at least a plurality of scores has a score above a threshold value.

Example 52. The method of any one of embodiments 7-51, wherein determining the tissue and / or cell type that produces cfDNA as a function of a plurality of genomic locations of at least a portion of the cfDNA fragment endpoint comprises contacting the cfDNA fragment And comparing the Fourier transform of a plurality of genomic locations of at least a portion of the endpoints or a mathematical transformation thereof with a reference map.

[0215] Example 53. The method of any one of embodiments 7-32, further comprising generating a report comprising a list of determined tissues and / or cell types that produce isolated cfDNA.

Example 54. Isolation of cell-free DNA (cfDNA) from a biological sample from a subject, wherein the isolated cfDNA comprises a plurality of cfDNA fragments;

Determining a sequence associated with at least a portion of a plurality of cfDNA fragments;

determining the genomic location in the reference genome for at least a portion of the cfDNA fragment endpoints of the plurality of cfDNA fragments as a function of the cfDNA fragment sequence;

determining at least a portion of the tissue and / or cell type that produces cfDNA as a function of genomic location of at least a portion of the cfDNA fragment endpoint; And

identifying the disease or disorder as a function of the determined tissue and / or cell type that produces cfDNA

Gt; a &lt; / RTI &gt; disease or disorder in a subject.

Example 55. The method of embodiment 54, wherein determining the tissue and / or cell type that produces cfDNA comprises comparing the genomic location of at least a portion of the cfDNA fragment endpoint to one or more reference maps.

Example 56. The method of embodiment 54 or 55 wherein the step of determining the tissue and / or cell type producing the cfDNA comprises performing a mathematical transformation on the distribution of the genomic location of at least a portion of the plurality of cfDNA fragment endpoints How to include.

[0099] Embodiment 57. The method of embodiment 56, wherein the mathematical transformation comprises a Fourier transform.

[0323] Embodiment 58. The method as in any of the embodiments 54-57, further comprising determining a score for each of the coordinates of at least a portion of the reference genome, wherein the score comprises at least a plurality of cfDNA fragment ends and Wherein determining the tissue and / or cell type that produces the observed cfDNA fragment is determined as a function of its genomic location, and wherein determining at least a portion of the cell type comprises comparing the score to the one or more reference maps.

[0215] Embodiment 59. The method of embodiment 58 wherein the score for the coordinates indicates or is related to the probability that the coordinates are at the location of the cfDNA fragment end point.

Example 60. The method as in any of the embodiments 55-59, wherein the reference map is obtained from a sample obtained from at least one reference subject and the DNase corresponding to at least one cell type or tissue associated with the disease or disorder I hypersensitive site data set, RNA expression data set, expression data, chromosome stereotypic map, chromatin accessibility map, chromatin fragmentation map or sequence data, and / or the location of nucleosomes and / Gt;

Example 61. The method of any one of embodiments 55-60, wherein the reference map is generated by digesting chromatin from at least one cell type or tissue with an exogenous nuclease (micrococus nuclease) How it is.

Example 62. The method of embodiment 60 or example 61, wherein the reference map comprises chromatin accessibility data determined by applying a potential-based method (e. G., ATAC-seq) to nuclei or chromatin from at least one cell type or tissue &Lt; / RTI &gt;

[0215] Example 63. The method as in any of the embodiments 55-62, wherein the reference map comprises data associated with DNA binding to the tissue or cell type and / or the location of the DNA occupancy protein.

Example 64. The method of embodiment 63, wherein the DNA binding and / or DNA occupancy protein is a transcription factor.

Example 65. The method of embodiment 63 or example 64 wherein the chromatin immunoprecipitation of the cross-linked DNA-protein complex is determined by applying to at least one cell type or tissue.

Example 66. The method of embodiment 63 or example 64 wherein the location is determined by treating the DNA associated with the tissue or cell type with a nuclease (e. G., DNase-I).

Example 67. The method of any one of embodiments 54-66, wherein the reference map is indicative of a biological feature associated with the location or spacing of a nucleosome, a chromatome, or other DNA binding or DNA occupancy protein in a tissue or cell type &Lt; / RTI &gt;

Example 68. The method of embodiment 67, wherein the biological characteristic is quantitative expression of one or more genes.

Example 69. The method of embodiment 67 or 68 wherein the biological characteristic is the presence or absence of one or more histone marks.

Example 70. The method of any one of embodiments 67-69, wherein the biological characteristic is hypersensitive to nuclease excision.

[0080] Embodiment 71. The method as in any of the embodiments 55-70, wherein the tissue or cell type used to create the reference map is a primary tissue from a subject having a disease or disorder.

Example 72. The method of embodiment 71 wherein the disease or disorder is associated with cancer, normal pregnancy, complication of pregnancy (e. G., Isomeric pregnancy), myocardial infarction, inflammatory bowel disease, systemic autoimmune disease, local autoimmune disease, Allogeneic transplantation, allografts without rejection, stroke, and local tissue injury.

[0216] 73. The method as in any of the embodiments 55-70, wherein the tissue or cell type used to generate the reference map is a primary tissue from a healthy subject.

[0213] 74. The method as in any of the embodiments 55-70, wherein the tissue or cell type used to create the reference map is an immortalized cell line.

[0214] [0094] Embodiment 75. The method as in any of the embodiments 55-70, wherein the tissue or cell type used to create the reference map is a biopsy from the tumor.

Example 76. The method of embodiment 60, wherein the sequence data obtained from the sample obtained from at least one reference object comprises the location of the cfDNA fragment end point probability.

[0213] 77. The method of embodiment 76, wherein the reference object is a healthy subject.

[0216] 78. The method of embodiment 76, wherein the reference entity has a disease or disorder.

Example 79. The method of embodiment 78 wherein the disease or disorder is associated with cancer, normal pregnancy, complication of pregnancy (e. G., Isomeric pregnancy), myocardial infarction, inflammatory bowel disease, systemic autoimmune disease, local autoimmune disease, Allogeneic transplantation, allografts without rejection, stroke, and local tissue injury.

[0322] 94. The method of any one of embodiments 60-81, wherein the reference map comprises a cfDNA fragment endpoint probability for at least a portion of the reference genome associated with the tissue or cell type.

[0215] Embodiment 81. The method of embodiment 80, wherein the reference map comprises a mathematical transformation of the cfDNA fragment endpoint probability.

Example 82. The method of embodiment 80, wherein the cfDNA fragment endpoint probability represents a subset of all reference genomic coordinates for tissue or cell type.

[0215] Example 83. The method of embodiment 82, wherein the subset is associated with the position or spacing of the nucleosome and / or the chromatome.

Example 84. The method of embodiment 82 or 83 wherein the subset is associated with a transcription initiation site and / or a transcription termination site.

[0324] 89. The method as in any of the embodiments 82-84, wherein the subset is associated with a binding site of at least one transcription factor.

[0324] 86. The method as in any of the embodiments 82-85, wherein the subset is associated with a nuclease hypersensitive region.

[0099] Embodiment 87. The method as in any of the embodiments 82-86, wherein the subset is further associated with at least one orthogonal biological feature.

[0454] 88. The method of embodiment 87 wherein the orthogonal biological features are associated with a high expression gene.

Example 89. The method of embodiment 87, wherein orthogonal biological features are associated with a low expression gene.

[0086] 90. The method as in any of the embodiments 81-89, wherein the mathematical transform includes a Fourier transform.

[0214] 91. The method as in any one of embodiments 58-90, wherein a subset of at least a plurality of cfDNA fragment endpoint scores each have a score that is greater than a threshold value.

Example 92. The method of any one of embodiments 54-91, wherein the tissue (s) and / or cell type (s) of cfDNA is determined as a function of a plurality of genomic locations of at least a portion of a cfDNA fragment endpoint Wherein the step comprises comparing a Fourier transform of a plurality of genomic locations of at least a portion of a cfDNA fragment endpoint, or a mathematical transformation thereof, with a reference map.

[0323] Embodiment 93. The method as in any of the embodiments 54-92, wherein the reference map comprises DNA or chromatin fragmentation data corresponding to at least one tissue associated with the disease or disorder.

[0213] 94. The method as in any of the examples 54-93, wherein the reference genome is human.

[0323] 95. The method as in any of the embodiments 54-94, further comprising generating a report that includes a statement identifying the disease or disorder.

Example 96. The method of embodiment 95, wherein the report further comprises a list of determined tissue (s) and / or cell type (s) of isolated cfDNA.

[0322] 98. The method of any one of embodiments 7-6, wherein the biological sample comprises, consists essentially of, or consists of whole blood, peripheral blood plasma, urine or cerebrospinal fluid.

(A), (b) and / or (d) by constructing a library of cfDNA and massively parallel sequencing, and (ii) obtaining a biological sample from the subject, c) to prepare a nucleosome map;

(a), (b) and / or (c) by obtaining a biological sample from a control subject or a subject with a known disease, isolating the cfDNA from the biological sample, and constructing the library of cfDNA and mass- To produce a reference set of nucleosome maps; And

(iii) comparing the nucleosome map derived from cfDNA to a reference set of nucleosome maps to determine the tissue and / or cell type that produces cfDNA

; here

(a) is the distribution of the likelihood that any particular base pair within the human genome will appear at the end of the cfDNA fragment;

(b) is the distribution of the likelihood that any pair of base pairs of the human genome will appear as the ends of a pair of cfDNA fragments;

(c) is a distribution of the likelihood that any particular base pair in the human genome will appear within the cfDNA fragment as a result of differentially occupied nucleosomes.

A method for determining the tissue and / or cell type that produces cfDNA in a subject.

Example 99. (i) The distribution (a), (b) and / or (c) was measured by obtaining a biological sample from a subject, isolating cfDNA from the biological sample, constructing a library of cfDNA and massively parallel sequencing Creating a nucleosome map;

(ii) obtaining a biological sample from a control subject or a subject having a known disease, isolating cfDNA from the biological sample, digesting chromatin with a micrococycle nuclease (MNase), DNase treatment or ATAC-Seq Measuring distribution (a), (b) and / or (c) by library construction and massively parallel sequencing to create a reference set of nucleosome maps; And

(iii) comparing the nucleosome map derived from cfDNA to a reference set of nucleosome maps to determine the tissue and / or cell type that produces cfDNA

; here

(a) is the distribution of the likelihood that any particular base pair within the human genome will appear at the end of the sequenced fragment;

(b) is a distribution of the likelihood that any pair of base pairs of the human genome will appear as a pair of ends of the sequenced fragments;

(c) is a distribution of the likelihood that any particular base pair in the human genome will appear within the sequenced fragment as a result of differential nucleosomal occupancy.

A method for determining the tissue and / or cell type that produces acellular DNA in a subject.

(A), (b) and / or (c) were determined by obtaining a biological sample from a subject, isolating cfDNA from the biological sample, constructing a library of cfDNA, and massively parallel sequencing Creating a nucleosome map;

(a), (b) and / or (c) by obtaining a biological sample from a control subject or a subject with a known disease, isolating the cfDNA from the biological sample, and constructing the library of cfDNA and mass- To produce a reference set of nucleosome maps; And

(iii) comparing the nucleosome map derived from cfDNA to a reference set of nucleosome maps to determine the clinical condition

; here

(a) is the distribution of the likelihood that any particular base pair within the human genome will appear at the end of the cfDNA fragment;

(b) is the distribution of the likelihood that any pair of base pairs of the human genome will appear as the ends of a pair of cfDNA fragments;

(c) is a distribution of the likelihood that any particular base pair in the human genome will appear within the cfDNA fragment as a result of differentially occupied nucleosomes.

A method of diagnosing a clinical condition in a subject.

Example 101.

(i) measuring distribution (a), (b) and / or (c) by obtaining a biological sample from a subject, isolating cfDNA from the biological sample, constructing a library of cfDNA and massively parallelizing the sequence, ;

(ii) obtaining a biological sample from a control subject or a subject having a known disease, isolating cfDNA from the biological sample, digesting chromatin with a micrococycle nuclease (MNase), DNase treatment or ATAC-Seq Measuring distribution (a), (b) and / or (c) by library construction and massively parallel sequencing to create a reference set of nucleosome maps; And

(iii) comparing the nucleosome map derived from cfDNA to the reference set of nucleosome maps to determine the original tissue composition of cfDNA

; here

(a) is the distribution of the likelihood that any particular base pair within the human genome will appear at the end of the sequenced fragment;

(b) is a distribution of the likelihood that any pair of base pairs of the human genome will appear as a pair of ends of the sequenced fragments;

(c) is a distribution of the likelihood that any particular base pair in the human genome will appear within the sequenced fragment as a result of differential nucleosomal occupancy.

A method of diagnosing a clinical condition in a subject.

Example 102. The method of any one of embodiments 98- 101, wherein a nucleosome map is

Purifying the isolated cfDNA from the biological sample;

Constructing the library by adapter ligation and optionally PCR amplification; And

Sequencing the generated library

. &Lt; / RTI &gt;

Example 103. In any one of the embodiments 98-101, the reference set of nucleosome maps is

Purifying the isolated cfDNA from the biological sample from the control subject;

Constructing the library by adapter ligation and optionally PCR amplification; And

Sequencing the generated library

. &Lt; / RTI &gt;

[0061] Embodiment 104. The method as in any of the embodiments 98-101, wherein after the mathematical transformation of one of the distributions (a), (b), or (c), or one of these distributions is applied to the Fourier transform in the adjacent window, Wherein quantification of the intensity over a frequency range associated with the occupancy of the nucleosomes is performed to summarize the degree to which the nucleosome represents a structured arrangement within each adjacent window.

Embodiment 105. The method as in any of the embodiments 98-101, wherein in a mathematical transformation of one of distributions (a), (b), or (c) (TFBS) is coupled by TF to summarize the nucleosome placement as a result of transcription factor (TF) activity in the immediate vicinity of the TFBS of the specific TF located immediately on the side of the nucleosome A method for quantifying the distribution of a region in a reference human genome to which a sequence reading initiation site is mapped.

Example 106. The method of any one of embodiments 98- 101, wherein the nucleosomal occupancy signal is derived from another genomic landmark such as a DNase I hypersensitive site, a transcription initiation site, a phase domain, (A), (b) and / or (c), or a mathematical transformation of one of these distributions around a subset of all such sites defined by a set of correlated behaviors (e.g., gene expression, etc.) And summarized according to any one of the aggregated signals.

Embodiment 107. The method as in any one of embodiments 98- 101, wherein the distribution is selected from the group consisting of, for example, a proximity window, or alternatively, a transcription factor binding site, Such as quantifying the periodicity in a discontinuous subset of the genome defined by the tissue expression data or other correlates of the nucleosome arrangement, or by measuring the periodic signal of the nucleosome arrangement within the various subset of the genome, Lt; / RTI &gt;

[0213] 108. The method as in any one of embodiments 98-101, wherein the distribution is defined by tissue-specific data, i.e., aggregated signals near tissue-specific DNase I hypersensitive sites.

[0303] 109. The method as in any one of embodiments 98-101, further comprising a statistical signal processing step for comparing the additional nucleosome map (s) to a reference set.

Example 110. In Example 109, a long range of nucleosome sequences was first summarized in a continuous window along the genomes of the various sample sets, followed by principal component analysis (PCA) to cluster the samples or estimate the mixture ratio How to do it.

[0215] Embodiment 111. The method of embodiment 100 or embodiment 101, wherein the clinical condition is cancer, a malignant tumor.

[0215] Embodiment 112. The method of embodiment 111, wherein the biological sample is circulating plasma, wherein a portion of the biological sample is derived from the tumor.

Example 113. The method of embodiment 100 or 101 wherein the clinical condition is selected from the group consisting of tissue damage, myocardial infarction (acute injury of cardiac tissue), autoimmune disease (chronic injury of various tissues), pregnancy, Trisomy) and rejection of transplantation.

[0324] 114. The method as in any of the embodiments 7-13, further comprising assigning a ratio for each of at least one tissue or cell type determined to contribute to the cfDNA.

Example 115. The method of embodiment 114 wherein the ratio assigned to each of the one or more determined tissues or cell types is based at least in part on the degree of correlation or increased correlation relative to cfDNA from healthy subjects or subjects Way.

Example 116. The method of embodiment 114 or example 115, wherein the degree of correlation is based, at least in part, on a comparison of the mathematical transformation of the distribution of the cfDNA fragment endpoints from the biological sample and the reference map associated with the determined tissue or cell type / RTI &gt;

[045] Example 117. The method as in any one of embodiments 114-116, wherein the proportion assigned to each of the one or more determined tissues or cell types is based on a mixture model.

From the foregoing, it will be understood that, although specific embodiments of the invention have been described herein for purposes of illustration, various modifications may be made without departing from the scope of the invention. Accordingly, the invention is not to be limited except as by the appended claims.

                         SEQUENCE LISTING <110> SHENDURE, Jay   <120> METHODS OF DETERMINING TISSUES AND / OR CELL TYPES GIVING RISE TO        CELL-FREE DNA, AND METHODS OF IDENTIFYING DISEASE OR DISORDER        USING SAME <130> 72227-8115.WO00 <140> PCT / US2015 / 042310 <141> 2015-07-27 <150> US 62 / 029,178 <151> 2014-07-25 <150> US 62 / 087,619 <151> 2014-12-04 <160> 4 <170> PatentIn version 3.5 <210> 1 <211> 34 <212> DNA <213> Artificial Sequence <220> Synthetic oligonucleotide: CL9 <400> 1 gtgactggag ttcagacgtg tgctcttccg atct 34 <210> 2 <211> 16 <212> DNA <213> Artificial Sequence <220> Synthetic oligonucleotide: Adapter 2.1 <220> <221> misc_feature &Lt; 222 > (16) <223> ddT at 3 'end <400> 2 cgacgctctt ccgatc 16 <210> 3 <211> 34 <212> DNA <213> Artificial Sequence <220> Synthetic oligonucleotide: Adapter 2.2 <220> <221> misc_feature <222> (1) <223> 5Phos at 5 'end <220> <221> misc_feature &Lt; 222 &gt; (30) <223> Phosphorothioate bond <220> <221> misc_feature &Lt; 222 > (31) <223> Phosphorothioate bond <220> <221> misc_feature &Lt; 222 > (32) <223> Phosphorothioate bond <220> <221> misc_feature &Lt; 222 > (33) <223> Phosphorothioate bond <400> 3 agatcggaag agcgtcgtgt agggaaagag tgta 34 <210> 4 <211> 10 <212> DNA <213> Artificial Sequence <220> Synthetic oligonucleotide: CL78 <220> <221> misc_feature <222> (1) <223> 5Phos at 5 'end <220> <221> misc_feature &Lt; 222 &gt; (10) (ISpC3) 10 and 3BioTEG at 3 'end <400> 4 agatcggaag 10

Claims (111)

Isolating cell-free DNA (cfDNA) from a biological sample from a subject, wherein the isolated cfDNA comprises a plurality of cfDNA fragments;
Determining a sequence associated with at least a portion of a plurality of cfDNA fragments;
determining the genomic location in the reference genome for at least a portion of the cfDNA fragment endpoints of the plurality of cfDNA fragments as a function of the cfDNA fragment sequence; And
determining at least a portion of the tissue and / or cell type that produces the cfDNA fragment as a function of the genomic location of at least a portion of the cfDNA fragment endpoint
Gt; and / or &lt; / RTI &gt; the cell type that produces the cfDNA in the subject. 2. The method of claim 1, wherein determining at least a portion of the tissue and / or cell type that produces the cfDNA fragment comprises comparing at least a portion of the genomic location of the cfDNA fragment endpoint to one or more reference maps. 3. The method of claim 1 or 2, wherein determining at least a portion of the tissue and / or cell type that produces the cfDNA fragment comprises performing a mathematical transformation on the distribution of the genomic location of at least a portion of the cfDNA fragment endpoint How it is. 4. The method of claim 3, wherein the mathematical transform comprises a Fourier transform. 5. The method of any one of claims 1 to 4, further comprising determining a score for each of at least some coordinates of the reference genome, wherein the score is a function of at least a plurality of cfDNA fragment endpoints and their genomic locations Determining and determining at least a portion of the tissue and / or cell type that produces the observed cfDNA fragment comprises comparing the score with one or more reference maps. 6. The method of claim 5, wherein the score for the coordinates indicates or is related to the probability that the coordinates are the positions of the cfDNA fragment endpoints. 7. The method according to any one of claims 2 to 6, wherein the reference map comprises a DNase I hypersensitive site data set generated from at least one cell type or tissue. 8. The method according to any one of claims 2 to 7, wherein the reference map comprises an RNA expression data set generated from at least one cell type or tissue. 9. The method according to any one of claims 2 to 8, wherein the reference map is generated from cfDNA from an animal xenograft or cells xenografted. 10. The method according to any one of claims 2 to 9, wherein the reference map comprises a chromosome stereochemistry map created from at least one cell type or tissue. 11. A method according to any one of claims 2 to 10, wherein the reference map comprises a chromatic accessibility map made from at least one cell type or tissue. 12. The method according to any one of claims 2 to 11, wherein the reference map comprises sequence data obtained from a sample obtained from at least one reference object. 13. The method according to any one of claims 2 to 12, wherein the reference map corresponds to at least one cell type or tissue associated with the disease or disorder. 14. The method according to any one of claims 2 to 13, wherein the reference map comprises the location or spacing of nucleosomes and / or chromatomes within a tissue or cell type. 15. The method according to any one of claims 2 to 14, wherein the reference map is generated by digesting a chromatin obtained from at least one cell type or tissue with an exogenous nuclease (e. G., Micrococus nuclease) How it is. 16. The method according to any one of claims 2 to 15, wherein the reference map comprises chromatin accessibility data determined by a displacement-based method (e.g., ATAC-seq). 17. The method according to any one of claims 2 to 16, wherein the reference map comprises data associated with DNA binding and / or the location of a DNA occupancy protein for a tissue or cell type. 18. The method of claim 17, wherein the DNA binding and / or DNA occupancy protein is a transcription factor. 19. The method according to claim 17 or 18, wherein the position is determined by chromatin immunoprecipitation of the cross-linked DNA-protein complex. 19. The method according to claim 17 or 18, wherein the location is determined by treating the DNA associated with the tissue or cell type with a nuclease (e.g., DNase-I). 21. The method according to any one of claims 2 to 20, wherein the reference map comprises a biological characteristic associated with the position or spacing of a nucleosome, a chromatome, or other DNA binding or DNA occupancy protein in a tissue or cell type Way. 22. The method of claim 21, wherein the biological characteristic is quantitative expression of one or more genes. 23. The method of claim 21 or 22, wherein the biological feature is the presence or absence of one or more histone marks. 24. The method according to any one of claims 21 to 23, wherein the biological characteristic is hypersensitive to nuclease cleavage. 25. The method according to any one of claims 2 to 24, wherein the tissue or cell type used to create the reference map is a primary tissue from a subject having a disease or disorder. 26. The method of claim 25 wherein the disease or disorder is selected from the group consisting of cancer, normal pregnancy, complications of pregnancy (e. G., Ischemic pregnancy), myocardial infarction, inflammatory bowel disease, systemic autoimmune disease, local autoimmune disease, Allograft, non-rejection, allograft, stroke, and local tissue injury. 25. The method according to any one of claims 2 to 24, wherein the tissue or cell type used to create the reference map is a primary tissue from a healthy subject. 25. The method according to any one of claims 2 to 24, wherein the tissue or cell type used to prepare the reference map is an immortalized cell line. 25. The method according to any one of claims 2 to 24, wherein the tissue or cell type used to create the reference map is a biopsy from the tumor. 13. The method of claim 12, wherein the sequence data comprises the position of the cfDNA fragment end point. 31. The method of claim 30, wherein the reference object is a healthy object. 31. The method of claim 30, wherein the reference entity has a disease or disorder. 33. The method of claim 32, wherein the disease or disorder is selected from the group consisting of cancer, normal pregnancy, complication of pregnancy (e. G., Isomeric pregnancy), myocardial infarction, inflammatory bowel disease, systemic autoimmune disease, local autoimmune disease, Allograft, non-rejection, allograft, stroke, and local tissue injury. 34. The method according to any one of claims 13 to 33, wherein the reference map comprises a reference score for at least a portion of the coordinates of the reference genome associated with the tissue or cell type. 35. The method of claim 34, wherein the reference map comprises a mathematical transformation of the score. 35. The method of claim 34, wherein the score represents a subset of all reference genomic coordinates for tissue or cell type. 37. The method of claim 36, wherein the subset is associated with the position or spacing of the nucleosome and / or the chromatome. 37. The method of claim 36 or 37, wherein the subset is associated with a transcription initiation site and / or a transcription termination site. 39. The method according to any one of claims 36 to 38, wherein the subset is associated with a binding site of at least one transcription factor. 40. The method according to any one of claims 36 to 39, wherein the subset is associated with a nuclease hypersensitive site. 40. The method according to any one of claims 36 to 40, wherein the subset is further associated with at least one orthogonal biological feature. 42. The method of claim 41, wherein the orthogonal biological feature is associated with a high expression gene. 42. The method of claim 41, wherein the orthogonal biological features are associated with a low expression gene. 44. The method of any one of claims 35 to 43, wherein the mathematical transform comprises a Fourier transform. 45. The method of any one of claims 5-44, wherein the subset of at least a plurality of scores has a score above a threshold value. 45. The method according to any one of claims 1 to 45, wherein determining the tissue and / or cell type that produces cfDNA as a function of a plurality of genomic locations of at least a portion of the cfDNA fragment endpoint comprises determining And comparing the Fourier transform of the plurality of genomic locations or a mathematical transformation thereof with a reference map. 47. The method of any one of claims 1 to 46, further comprising generating a report comprising a list of determined tissues and / or cell types that produce isolated cfDNA. Isolating cell-free DNA (cfDNA) from a biological sample from a subject, wherein the isolated cfDNA comprises a plurality of cfDNA fragments;
Determining a sequence associated with at least a portion of a plurality of cfDNA fragments;
determining the genomic location in the reference genome for at least a portion of the cfDNA fragment endpoints of the plurality of cfDNA fragments as a function of the cfDNA fragment sequence;
determining at least a portion of the tissue and / or cell type that produces cfDNA as a function of genomic location of at least a portion of the cfDNA fragment endpoint; And
identifying the disease or disorder as a function of the determined tissue and / or cell type that produces cfDNA
Gt; a &lt; / RTI &gt; disease or disorder in a subject. 49. The method of claim 48, wherein determining the tissue and / or cell type that produces cfDNA comprises comparing the genomic location of at least a portion of the cfDNA fragment endpoint to one or more reference maps. 49. The method of claim 48 or 49, wherein determining the tissue and / or cell type that produces cfDNA comprises performing a mathematical transformation on the distribution of genomic locations of at least a portion of the plurality of cfDNA fragment endpoints Way. 51. The method of claim 50, wherein the mathematical transform comprises a Fourier transform. 51. The method of any one of claims 48 to 51, further comprising determining a score for each of the coordinates of at least a portion of the reference genome, wherein the score is a function of at least a plurality of cfDNA fragment endpoints and their genomic locations , And determining at least a portion of the tissue and / or cell type that produces the observed cfDNA fragment comprises comparing the score to one or more reference maps. 53. The method of claim 52, wherein the score for the coordinates indicates or is associated with a probability that the coordinates are positions of the cfDNA fragment endpoint. 53. A method according to any one of claims 49 to 53, wherein the reference map is obtained from a sample obtained from at least one reference subject and comprises a DNase I hypersensitive site data set corresponding to at least one cell type or tissue associated with the disease or disorder , RNA expression data sets, expression data, chromosome stereotactic maps, chromosome accessibility maps, chromosomal fragmentation maps or sequence data, and / or locations or intervals of nucleosomes and / or chromatomes within tissue or cell types / RTI &gt; 54. The method according to any one of claims 49 to 54, wherein the reference map is generated by digesting chromatin from at least one cell type or tissue with an exogenous nuclease (e. G., Micrococus nuclease) / RTI &gt; 56. The method of claim 54 or 55, wherein the reference map comprises chromatin accessibility data determined by applying a potential-based method (e.g., ATAC-seq) to nuclei or chromatin from at least one cell type or tissue / RTI &gt; 57. The method of any one of claims 49-56, wherein the reference map comprises data associated with DNA binding to the tissue or cell type and / or the location of the DNA occupancy protein. 58. The method of claim 57, wherein the DNA binding and / or DNA occupancy protein is a transcription factor. 59. The method of claim 57 or 58, wherein the chromatin immune precipitation of the cross-linked DNA-protein complex is determined by applying to at least one cell type or tissue. 59. The method of claim 57 or 58, wherein the location is determined by treating DNA associated with the tissue or cell type with a nuclease (e. G., DNase-I). 60. The method according to any one of claims 48 to 60, wherein the reference map comprises a biological feature associated with the location or spacing of a nucleosome, a chromatome, or other DNA binding or DNA occupancy protein in a tissue or cell type Way. 62. The method of claim 61, wherein the biological characteristic is quantitative expression of one or more genes. 62. The method of claim 61 or 62, wherein the biological feature is the presence or absence of one or more histone marks. 63. The method according to any one of claims 61 to 63, wherein the biological characteristic is hypersensitive to nuclease excision. 65. The method according to any one of claims 49 to 64, wherein the tissue or cell type used to create the reference map is a primary tissue from a subject having a disease or disorder. 65. The method of claim 65, wherein the disease or disorder is selected from the group consisting of cancer, normal pregnancy, complication of pregnancy (e. G., Ischemic pregnancy), myocardial infarction, inflammatory bowel disease, systemic autoimmune disease, local autoimmune disease, Allograft, non-rejection, allograft, stroke, and local tissue injury. 65. The method according to any one of claims 49 to 65, wherein the tissue or cell type used to create the reference map is a primary tissue from a healthy subject. 65. The method according to any one of claims 49 to 65, wherein the tissue or cell type used to create the reference map is an immortalized cell line. 65. The method according to any one of claims 49 to 65, wherein the tissue or cell type used to create the reference map is a biopsy from the tumor. 55. The method of claim 54, wherein the sequence data obtained from the sample obtained from at least one reference object comprises the location of the cfDNA fragment endpoint probability. 70. The method of claim 70, wherein the reference object is a healthy object. 71. The method of claim 70, wherein the reference entity has a disease or disorder. 73. The method of claim 72, wherein the disease or disorder is selected from the group consisting of cancer, normal pregnancy, complications of pregnancy (e. G., Ischemic pregnancy), myocardial infarction, inflammatory bowel disease, systemic autoimmune disease, Allograft, non-rejection, allograft, stroke, and local tissue injury. 73. The method of any one of claims 54 to 73, wherein the reference map comprises a cfDNA fragment endpoint probability for at least a portion of the reference genome associated with the tissue or cell type. 75. The method of claim 74, wherein the reference map comprises a mathematical transformation of the cfDNA fragment endpoint probability. 74. The method of claim 74, wherein the cfDNA fragment endpoint probability represents a subset of all reference genomic coordinates for tissue or cell type. 78. The method of claim 76, wherein the subset is associated with the position or spacing of the nucleosome and / or the chromatome. 78. The method of claim 76 or claim 77, wherein the subset is associated with a transcription initiation site and / or a transcription termination site. 78. The method of any one of claims 76-78, wherein the subset is associated with a binding site of at least one transcription factor. 80. The method according to any one of claims 76 to 79, wherein the subset is associated with a nuclease hypersensitive site. 80. The method according to any one of claims 76 to 80, wherein the subset is further associated with at least one orthogonal biological feature. 83. The method of claim 81, wherein the orthogonal biological feature is associated with a high expression gene. 83. The method of claim 81, wherein the orthogonal biological feature is associated with a low expression gene. 83. The method of any one of claims 75-83, wherein the mathematical transform comprises a Fourier transform. 84. The method of any one of claims 52-84, wherein a subset of at least a plurality of cfDNA fragment endpoint scores each have a score above a threshold value. 83. The method according to any one of claims 48 to 85, wherein the step of determining tissue (s) and / or cell type (s) producing cfDNA as a function of a plurality of genomic locations of at least a portion of a cfDNA fragment end- And comparing the Fourier transform of a plurality of genomic locations of at least a portion of the fragment endpoint or a mathematical transformation thereof with a reference map. 87. The method according to any one of claims 48 to 86, wherein the reference map comprises DNA or chromatin fragmentation data corresponding to at least one tissue associated with the disease or disorder. 87. The method according to any one of claims 48 to 87, wherein the reference genome is human. 90. The method of any one of claims 48 to 88, further comprising creating a report that includes a statement identifying the disease or disorder. 90. The method of claim 89, wherein the report further comprises a list of determined tissue (s) and / or cell type (s) of isolated cfDNA. 90. The method according to any one of claims 1 to 90, wherein the biological sample comprises, consists essentially of, or consists of whole blood, peripheral blood plasma, urine or cerebrospinal fluid. (a), (b) and / or (c) by (i) obtaining a biological sample from a subject, isolating cell-free DNA (cfDNA) from the biological sample, and constructing a library of cfDNA and massively parallel sequencing Creating a nucleosome map;
(a), (b) and / or (c) by obtaining a biological sample from a control subject or a subject with a known disease, isolating the cfDNA from the biological sample, and constructing the library of cfDNA and mass- To produce a reference set of nucleosome maps; And
(iii) comparing the nucleosome map derived from cfDNA to a reference set of nucleosome maps to determine the tissue and / or cell type that produces cfDNA
; here
(a) is the distribution of the likelihood that any particular base pair within the human genome will appear at the end of the cfDNA fragment;
(b) is the distribution of the likelihood that any pair of base pairs of the human genome will appear as the ends of a pair of cfDNA fragments;
(c) is a distribution of the likelihood that any particular base pair in the human genome will appear within the cfDNA fragment as a result of differentially occupied nucleosomes.
A method for determining the tissue and / or cell type that produces cfDNA in a subject. (i) measuring distribution (a), (b) and / or (c) by obtaining a biological sample from a subject, isolating cfDNA from the biological sample, constructing a library of cfDNA and massively parallelizing the sequence, ;
(ii) obtaining a biological sample from a control subject or a subject having a known disease, isolating cfDNA from the biological sample, digesting chromatin with a micrococycle nuclease (MNase), DNase treatment or ATAC-Seq Measuring distribution (a), (b) and / or (c) by library construction and massively parallel sequencing to create a reference set of nucleosome maps; And
(iii) comparing the nucleosome map derived from cfDNA to a reference set of nucleosome maps to determine the tissue and / or cell type that produces cfDNA
; here
(a) is the distribution of the likelihood that any particular base pair within the human genome will appear at the end of the sequenced fragment;
(b) is a distribution of the likelihood that any pair of base pairs of the human genome will appear as a pair of ends of the sequenced fragments;
(c) is a distribution of the likelihood that any particular base pair in the human genome will appear within the sequenced fragment as a result of differential nucleosomal occupancy.
A method for determining the tissue and / or cell type that produces acellular DNA in a subject. (i) measuring distribution (a), (b) and / or (c) by obtaining a biological sample from a subject, isolating cfDNA from the biological sample, constructing a library of cfDNA and massively parallelizing the sequence, ;
(a), (b) and / or (c) by obtaining a biological sample from a control subject or a subject with a known disease, isolating the cfDNA from the biological sample, and constructing the library of cfDNA and mass- To produce a reference set of nucleosome maps; And
(iii) comparing the nucleosome map derived from cfDNA to a reference set of nucleosome maps to determine the clinical condition
; here
(a) is the distribution of the likelihood that any particular base pair within the human genome will appear at the end of the cfDNA fragment;
(b) is the distribution of the likelihood that any pair of base pairs of the human genome will appear as the ends of a pair of cfDNA fragments;
(c) is a distribution of the likelihood that any particular base pair in the human genome will appear within the cfDNA fragment as a result of differentially occupied nucleosomes.
A method of diagnosing a clinical condition in a subject. (i) measuring distribution (a), (b) and / or (c) by obtaining a biological sample from a subject, isolating cfDNA from the biological sample, constructing a library of cfDNA and massively parallelizing the sequence, ;
(ii) obtaining a biological sample from a control subject or a subject having a known disease, isolating cfDNA from the biological sample, digesting chromatin with a micrococycle nuclease (MNase), DNase treatment or ATAC-Seq Measuring distribution (a), (b) and / or (c) by library construction and massively parallel sequencing to create a reference set of nucleosome maps; And
(iii) comparing the nucleosome map derived from cfDNA to the reference set of nucleosome maps to determine the original tissue composition of cfDNA
; here
(a) is the distribution of the likelihood that any particular base pair within the human genome will appear at the end of the sequenced fragment;
(b) is a distribution of the likelihood that any pair of base pairs of the human genome will appear as a pair of ends of the sequenced fragments;
(c) is a distribution of the likelihood that any particular base pair in the human genome will appear within the sequenced fragment as a result of differential nucleosomal occupancy.
A method of diagnosing a clinical condition in a subject. 95. The method according to any one of claims 92 to 95, wherein the nucleosome map is
Purifying the isolated cfDNA from the biological sample;
Constructing the library by adapter ligation and optionally PCR amplification; And
Sequencing the generated library
. &Lt; / RTI &gt; 95. The method according to any one of claims 92 to 95, wherein the reference set of nucleosome maps comprises
Purifying the isolated cfDNA from the biological sample from the control subject;
Constructing the library by adapter ligation and optionally PCR amplification; And
Sequencing the generated library
. &Lt; / RTI &gt; A method according to any one of claims 92 to 95, wherein after the mathematical transformation of one of the distributions (a), (b) or (c), or one of these distributions is applied to the Fourier transform in the adjacent window, Wherein a quantification of the intensity over a frequency range associated with the occupancy of the nucleosome is performed to summarize the degree of representation of the structured arrangement within the adjacent window. 95. The method according to any one of claims 92 to 95, wherein in the mathematical transformation of one of the distributions (a), (b) or (c) When the transcription factor binding site (TFBS) is joined by TF to summarize the nucleosome placement as a result of TF activity, the nucleosomes are often located immediately adjacent to the TFBS of the specific TF located immediately on the side of the nucleosome, A method for quantifying the distribution of sites within a reference human genome that is mapped. 95. The method according to any one of claims 92 to 95, wherein the nucleosomal occupancy signal is selected from the group consisting of correlated behavior in other genomic landmarks such as DNase I hypersensitive site, transcription initiation site, phase domain, (A), (b) and / or (c), or one of these distributions around a subset of all such regions as defined by a mathematical transformation One summarized by one. 95. The method according to any one of claims 92 to 95, wherein the distribution is selected from the group consisting of transcription factor binding sites, gene model features (e. G., Transcription initiation sites), tissue expression Data or a periodic signal of a nucleosome arrangement within the various subsets of the genome, such as quantifying the periodicity in a discontinuous subset of the genome defined by other correlates of the nucleosome arrangement How it is. 95. The method according to any one of claims 92 to 95, wherein the distribution is defined by tissue-specific data, i.e., aggregated signals near tissue-specific DNase I hypersensitive sites. 95. The method according to any one of claims 92 to 95, further comprising a statistical signal processing step for comparing the additional nucleosome map (s) to a reference set. 107. The method of claim 103, wherein first performing a principal component analysis (PCA) to summarize a long range of nucleosome sequences within a continuous window along the genomes of the various sample sets and then clustering the samples or estimating the mixture ratio. 95. The method of claim 94 or 95, wherein the clinical condition is cancer, i.e., a malignant tumor. 105. The method of claim 105, wherein the biological sample is circulating plasma, wherein the portion of the biological sample is cfDNA derived from the tumor. 95. The method of claim 94 or 95 wherein the condition is selected from the group consisting of tissue damage, myocardial infarction (acute injury of cardiac tissue), autoimmune disease (chronic injury of various tissues), pregnancy, chromosomal anomalies (e. And rejection of transplantation. 107. The method of any one of claims 1 to 107, further comprising assigning a ratio for each of at least one tissue or cell type determined to contribute to the cfDNA. 109. The method of claim 108 wherein the percentage assigned to each of the one or more determined tissues or cell types is based at least in part on the degree of correlation or increased correlation relative to cfDNA from healthy subjects or subjects. 109. The method of claim 108 or claim 109 wherein the degree of correlation is based at least in part on a mathematical transformation of the distribution of cfDNA fragment endpoints from the biological sample and a comparison of the reference map associated with the determined tissue or cell type. 112. The method according to any one of claims 108 to 110, wherein the proportion assigned to each of the one or more determined tissues or cell types is based on a mixture model.

Images